> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to SpecStream

The SpecStream connector lets your AI agent act in each user's SpecStream account. Each user connects their own SpecStream access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 235 tools.

**Authentication:** Bearer Token
**Categories:** Search, AI, CRM & Sales
**Tools:** 235: 125 read, 85 write, 25 destructive
**Users sign in with:** Access token
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=SPECSTREAM)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the SpecStream connection

   In **AgentKit > Connections**, create a SpecStream connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'specstream'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize SpecStream:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'specstream_get_account',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "specstream"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize SpecStream:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="specstream_get_account",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their SpecStream credentials there. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'specstream_agent_list_steps',
  toolInput: {},
  connector: 'specstream',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="specstream_agent_list_steps",
    tool_input={},
    connection_name="specstream",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `specstream_agent_list_steps`

Agent List Steps · Read-only

Read the buying steps and the ways to pay.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_browser_live_view`

Browser Live View · Read-only

Watch the browser's screen as it changes, in a web page.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for browser live view.
- `token` (`string`, required): The token in the session's live_url.

Example input: `{"id":"<id>","token":"<token>"}`

### `specstream_dictation_capabilities`

Dictation Capabilities · Read-only

Read configured hosted dictation support.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_export_account`

Export Account · Read-only

Account export.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_extract_pages`

Extract Pages · Read-only

Extract API: read up to 20 pages, live from their URLs or from HTML you send, each page's main text as Markdown or plain text, its title and its published date. Only a page read is charged.
With no API key, the call answers 402 with a PAYMENT-REQUIRED header (base64 x402 v2 PaymentRequired): pay for this one call with x402's `exact` scheme in USDC on Base, Arbitrum or Polygon, and send the same request again with the payment in PAYMENT-SIGNATURE. The reply then carries PAYMENT-RESPONSE naming the transaction. One search of up to 10 results costs $0.007 ($0.002 with speed=fast, news or images), and a page read costs $0.002 for each page you ask for. The price is taken when the payment settles, before the call runs, so a search with no result or a page with no text is paid. If the call fails on our side after the payment settled, or the settle has no answer yet, send the same request with the same PAYMENT-SIGNATURE again: it is answered, and nothing is taken twice.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `format` (`string`, optional): Format for extract pages. One of: markdown, text, html. One of: `markdown`, `text`, `html`. Default: `markdown`.
- `include` (`array`, optional): Include for extract pages.
- `pages` (`array`, optional): Pages for extract pages.
- `urls` (`array`, optional): Urls for extract pages.
- `x_end_user_cap_cents` (`integer`, optional): The most the end user named in X-End-User-Id may spend this UTC month, in cents. A call past it is refused with 402 and details.reason end_user_cap. Send it on each call; SpecStream stores no policy.
- `x_end_user_id` (`string`, optional): Your own label for the end user this call is for. SpecStream keeps only its SHA-256 hex, and GET /v1/account/usage?group=end_user sums your calls by it.

Example input: `{}`

### `specstream_find_events`

Find Events · Read-only

SSE stream with Last-Event-ID.
Needs id (identifiers returned by earlier SpecStream calls).
The response is an event stream returned as text: the first lines show the events received before the connection closes.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for find events.
- `after` (`integer`, optional): After for find events.
- `last_event_id` (`string`, optional): Optional value sent in the Last-Event-Id HTTP header.

Example input: `{"id":"<id>"}`

### `specstream_get_account`

Get Account · Read-only

Paid history.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_account_request`

Get Account Request · Read-only

Open one call of the request log, with its reply where it is kept.
Returns a JSON object with data.
Needs request_log_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `request_log_id` (`string`, required): Request Log ID for get account request.

Example input: `{"request_log_id":"<request_log_id>"}`

### `specstream_get_account_usage`

Get Account Usage · Read-only

Sum calls and spend by day, endpoint, key or end user, for the Search API, the Extract API and Deep Search, as JSON or CSV.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `format` (`string`, optional): csv returns text/csv, one line per group value and product. One of: json, csv. One of: `json`, `csv`. Default: `json`.
- `from` (`string`, optional): An ISO date or time, UTC when no zone is given.
- `group` (`string`, optional): end_user sums by the SHA-256 hex of each call's X-End-User-Id. One of: day, endpoint, key, end_user. One of: `day`, `endpoint`, `key`, `end_user`. Default: `day`.
- `to` (`string`, optional): An ISO date or time, not included; a date alone includes that day.

Example input: `{}`

### `specstream_get_agent_list_purchase`

Get Agent List Purchase · Read-only

Read a purchased list and its rows.
Returns a JSON object with purchase_id, status, request, reading, requested, found, and more.
Needs purchase_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchase_id` (`string`, required): The purchase_id from the 202.

Example input: `{"purchase_id":"<purchase_id>"}`

### `specstream_get_browser_session`

Get Browser Session · Read-only

Read a browser session: its status, why it ended, its seconds, actions, bytes and charge.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get browser session.

Example input: `{"id":"<id>"}`

### `specstream_get_capacity_auto_topup`

Get Capacity Auto Topup · Read-only

Read the automatic top-up setting, its saved card, this month's charges and the allowed ranges.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_balance`

Get Capacity Balance · Read-only

Read the available balance without approving new work.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, optional): Project ID for get capacity balance.

Example input: `{}`

### `specstream_get_capacity_catalog`

Get Capacity Catalog · Read-only

Read the current dollar balance and membership offer.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_entitlements`

Get Capacity Entitlements · Read-only

Read what the account's plan includes and how many members it uses.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, optional): Project ID for get capacity entitlements.

Example input: `{}`

### `specstream_get_capacity_execution`

Get Capacity Execution · Read-only

Read a find's approved work, final charge and your access to its results.
Returns a JSON object with data.
Needs find_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `find_id` (`string`, required): Find ID for get capacity execution.

Example input: `{"find_id":"<find_id>"}`

### `specstream_get_capacity_funding`

Get Capacity Funding · Read-only

Recover the original funding request from verified provider facts.
Returns a JSON object with data.
Needs funding_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `funding_id` (`string`, required): Funding ID for get capacity funding.

Example input: `{"funding_id":"<funding_id>"}`

### `specstream_get_capacity_history`

Get Capacity History · Read-only

Read account balance events including the reason for each bonus grant.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_membership`

Get Capacity Membership · Read-only

Read the current paid period and next-renewal choice.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_partner`

Get Capacity Partner · Read-only

Read a partner's own link, terms and monthly sign-ups, payments and commission.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_payment_receipt`

Get Capacity Payment Receipt · Read-only

Read one payment's receipt: paid, tax, credit added with any prepayment bonus, refunds and the cash value of the credit left; an app store payment answers 409 receipt_issued_by_store.
Returns a JSON object with data.
Needs payment_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `payment_id` (`string`, required): Payment ID for get capacity payment receipt.

Example input: `{"payment_id":"<payment_id>"}`

### `specstream_get_capacity_quote`

Get Capacity Quote · Read-only

Recover your signed quote and any find already approved from it.
Returns a JSON object with data.
Needs quote_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `quote_id` (`string`, required): Quote ID for get capacity quote.

Example input: `{"quote_id":"<quote_id>"}`

### `specstream_get_capacity_referral`

Get Capacity Referral · Read-only

Read the verified account's stable referral link and rewards.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_capacity_refund_claim`

Get Capacity Refund Claim · Read-only

Read the status of an owned correction claim.
Returns a JSON object with data.
Needs claim_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `claim_id` (`string`, required): Claim ID for get capacity refund claim.

Example input: `{"claim_id":"<claim_id>"}`

### `specstream_get_capacity_request`

Get Capacity Request · Read-only

Recover the find already started with this request key.
Returns a JSON object with data.
Needs request_key (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `request_key` (`string`, required): Request Key for get capacity request.
- `project_id` (`string`, optional): Project ID for get capacity request.

Example input: `{"request_key":"<request_key>"}`

### `specstream_get_capacity_request_rows`

Get Capacity Request Rows · Read-only

Read where the request asked with this key stands and the rows it found so far.
Returns a JSON object with data.
Needs request_key (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `request_key` (`string`, required): Request Key for get capacity request rows.
- `cursor` (`string`, optional): Cursor for get capacity request rows.
- `format` (`string`, optional): json returns the records. markdown or csv returns the asked columns in order as one table in data.table, with a source link for each value, in place of the records. One of: json, markdown, csv. One of: `json`, `markdown`, `csv`. Default: `json`.
- `limit` (`integer`, optional): Limit for get capacity request rows. Default: `100`.
- `project_id` (`string`, optional): Project ID for get capacity request rows.

Example input: `{"request_key":"<request_key>"}`

### `specstream_get_capacity_store_config`

Get Capacity Store Config · Read-only

Read the app store products and this account's store purchases.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_connection_columns`

Get Connection Columns · Read-only

Read destination columns before mapping fields.
Returns a JSON object with data.
Needs identity (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `identity` (`string`, required): Identity for get connection columns.
- `selection` (`object`, required): Selection for get connection columns.

Example input: `{"identity":"<identity>","selection":{}}`

### `specstream_get_crawl`

Get Crawl · Read-only

Read a crawl job: its state and its pages from skip; follow next_skip.
Returns a JSON object with data.
Needs crawl_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `crawl_id` (`string`, required): Crawl ID for get crawl.
- `skip` (`integer`, optional): The first page of results to return, as next_skip names it. Default: `0`.

Example input: `{"crawl_id":"<crawl_id>"}`

### `specstream_get_delivery`

Get Delivery · Read-only

Read your finished results with the request revision and sources behind them.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get delivery.

Example input: `{"id":"<id>"}`

### `specstream_get_find`

Get Find · Read-only

Owner find snapshot.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get find.

Example input: `{"id":"<id>"}`

### `specstream_get_find_commerce`

Get Find Commerce · Read-only

Read payment, fulfillment and access truth.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get find commerce.

Example input: `{"id":"<id>"}`

### `specstream_get_find_decisions`

Get Find Decisions · Read-only

Read currently authorized Match decisions.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get find decisions.
- `cursor` (`string`, optional): Cursor for get find decisions.
- `limit` (`integer`, optional): Limit for get find decisions. Default: `100`.

Example input: `{"id":"<id>"}`

### `specstream_get_find_lineage`

Get Find Lineage · Read-only

Read which request revision a find ran, and the find it continued.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get find lineage.

Example input: `{"id":"<id>"}`

### `specstream_get_find_rows`

Get Find Rows · Read-only

Read the delivered results and the source behind each condition.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get find rows.
- `cursor` (`string`, optional): Cursor for get find rows.
- `format` (`string`, optional): json returns the records. markdown or csv returns the asked columns in order as one table in data.table, with a source link for each value, in place of the records. One of: json, markdown, csv. One of: `json`, `markdown`, `csv`. Default: `json`.
- `limit` (`integer`, optional): Limit for get find rows. Default: `100`.

Example input: `{"id":"<id>"}`

### `specstream_get_investigation_batch`

Get Investigation Batch · Read-only

Read each input row, its search status and the list it led to.
Returns a JSON object with data.
Needs batch_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `batch_id` (`string`, required): Batch ID for get investigation batch.

Example input: `{"batch_id":"<batch_id>"}`

### `specstream_get_investigation_function`

Get Investigation Function · Read-only

Read the saved request and its typed inputs.
Returns a JSON object with data.
Needs function_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `function_id` (`string`, required): Function ID for get investigation function.

Example input: `{"function_id":"<function_id>"}`

### `specstream_get_investigation_inputs`

Get Investigation Inputs · Read-only

List the columns of a list with example values, to choose which column feeds each input.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for get investigation inputs.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_get_investigation_parameters`

Get Investigation Parameters · Read-only

List the values in a saved request that can become inputs.
Returns a JSON object with data.
Needs spec_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `spec_id` (`string`, required): Spec ID for get investigation parameters.
- `revision_id` (`string`, optional): Revision ID for get investigation parameters.

Example input: `{"spec_id":"<spec_id>"}`

### `specstream_get_investigation_results`

Get Investigation Results · Read-only

Read each input row with the results it led to, including their sources. Input values are never treated as sources for results.
Returns a JSON object with data.
Needs batch_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `batch_id` (`string`, required): Batch ID for get investigation results.

Example input: `{"batch_id":"<batch_id>"}`

### `specstream_get_me`

Get Me · Read-only

Spend ceiling and remaining_spend authorization state.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_get_project`

Get Project · Read-only

Read current project membership and authorization.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get project.

Example input: `{"id":"<id>"}`

### `specstream_get_reconciled_input`

Get Reconciled Input · Read-only

Read which chosen pair filled each cell of a list made by matching.
Returns a JSON object with data.
Needs import_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `import_id` (`string`, required): Import ID for get reconciled input.

Example input: `{"import_id":"<import_id>"}`

### `specstream_get_reconciliation`

Get Reconciliation · Read-only

Read a comparison: proposed pairs, alternatives, conflicts, missing values and sources.
Returns a JSON object with data.
Needs comparison_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `comparison_id` (`string`, required): Comparison ID for get reconciliation.

Example input: `{"comparison_id":"<comparison_id>"}`

### `specstream_get_reconciliation_fields`

Get Reconciliation Fields · Read-only

Read the fields of a list you can open that can be used to match it against another list.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for get reconciliation fields.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_get_research`

Get Research · Read-only

Read a research run: its state and progress, and its report when it is completed.
Returns a JSON object with data.
Needs run_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `run_id` (`string`, required): Run ID for get research.

Example input: `{"run_id":"<run_id>"}`

### `specstream_get_research_rule`

Get Research Rule · Read-only

Read a rule's limits, what it has spent, which rows are waiting and its recent runs.
Returns a JSON object with data.
Needs rule_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `rule_id` (`string`, required): Rule ID for get research rule.
- `offset` (`integer`, optional): Offset for get research rule. Default: `0`.

Example input: `{"rule_id":"<rule_id>"}`

### `specstream_get_research_rule_results`

Get Research Rule Results · Read-only

Read every run of the rule with its input rows and results.
Returns a JSON object with data.
Needs rule_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `rule_id` (`string`, required): Rule ID for get research rule results.

Example input: `{"rule_id":"<rule_id>"}`

### `specstream_get_revisit_report`

Get Revisit Report · Read-only

Read which saved results a refresh reached, and which it did not.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get revisit report.
- `after` (`string`, optional): After for get revisit report. Default: ``.
- `limit` (`integer`, optional): Limit for get revisit report. Default: `100`.

Example input: `{"id":"<id>"}`

### `specstream_get_shared_spec`

Get Shared Spec · Read-only

Read a public request snapshot while its issuer, access and link remain valid.
Returns a JSON object with data.
Needs token (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `token` (`string`, required): Token for get shared spec.

Example input: `{"token":"<token>"}`

### `specstream_get_spec`

Get Spec · Read-only

Read a saved request and its current revision.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get spec.

Example input: `{"id":"<id>"}`

### `specstream_get_spec_options`

Get Spec Options · Read-only

Read the conditions and details you can ask for on one record type.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `target` (`string`, required): Target for get spec options.

Example input: `{"target":"<target>"}`

### `specstream_get_spec_revision`

Get Spec Revision · Read-only

Read one revision of a saved request.
Returns a JSON object with data.
Needs id, revision_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for get spec revision.
- `revision_id` (`string`, required): Revision ID for get spec revision.

Example input: `{"id":"<id>","revision_id":"<revision_id>"}`

### `specstream_get_vocabulary`

Get Vocabulary · Read-only

Read the public bar vocabulary and ready-made details.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `if_none_match` (`string`, optional): Optional value sent in the If-None-Match HTTP header.

Example input: `{}`

### `specstream_get_web_source`

Get Web Source · Read-only

Read a saved website source and how each of its columns is read.
Returns a JSON object with data.
Needs source_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `source_id` (`string`, required): Source ID for get web source.

Example input: `{"source_id":"<source_id>"}`

### `specstream_get_web_source_preview`

Get Web Source Preview · Read-only

Read one source preview with its page hash, cell locations and field explanations.
Returns a JSON object with data.
Needs preview_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `preview_id` (`string`, required): Preview ID for get web source preview.

Example input: `{"preview_id":"<preview_id>"}`

### `specstream_list_account_keys`

List Account Keys · Read-only

List API keys.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_account_requests`

List Account Requests · Read-only

List the account's Search API and Extract API calls, charged or not, newest first.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for list account requests.
- `endpoint` (`string`, optional): Endpoint for list account requests. One of: /v1/web-search, /v1/web-search/batch, /v1/extract. One of: `/v1/web-search`, `/v1/web-search/batch`, `/v1/extract`.
- `from` (`string`, optional): An ISO date or time, UTC when no zone is given.
- `key_id` (`string`, optional): Key ID for list account requests.
- `limit` (`integer`, optional): Limit for list account requests. Default: `50`.
- `status` (`string`, optional): One status, such as 429, or a class, such as 4xx.
- `to` (`string`, optional): An ISO date or time, not included; a date alone includes that day.

Example input: `{}`

### `specstream_list_browser_sessions`

List Browser Sessions · Read-only

List the account's browser sessions, newest first.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for list browser sessions.
- `limit` (`integer`, optional): Limit for list browser sessions. Default: `20`.
- `status` (`string`, optional): Status for list browser sessions. One of: starting, running, ended. One of: `starting`, `running`, `ended`.

Example input: `{}`

### `specstream_list_capacity_occurrences`

List Capacity Occurrences · Read-only

Read completed, blocked, and missed occurrences.
Returns a JSON object with data.
Needs schedule_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `schedule_id` (`string`, required): Schedule ID for list capacity occurrences.
- `before` (`string`, optional): Before for list capacity occurrences.
- `limit` (`integer`, optional): Limit for list capacity occurrences. Default: `50`.

Example input: `{"schedule_id":"<schedule_id>"}`

### `specstream_list_capacity_payments`

List Capacity Payments · Read-only

List the account's payments as receipts, newest first; app store payments are counted, not listed, since the store issues their receipts.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for list capacity payments.
- `limit` (`integer`, optional): Limit for list capacity payments. Default: `50`.

Example input: `{}`

### `specstream_list_capacity_refund_claims`

List Capacity Refund Claims · Read-only

Read the account's correction claims, newest first.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_capacity_schedules`

List Capacity Schedules · Read-only

Read authorized recurring work and its current state.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): After for list capacity schedules. Default: ``.
- `limit` (`integer`, optional): Limit for list capacity schedules. Default: `50`.
- `project_id` (`string`, optional): Project ID for list capacity schedules.

Example input: `{}`

### `specstream_list_cloud_imports`

List Cloud Imports · Read-only

List daily source checks and import status.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_collection_investigations`

List Collection Investigations · Read-only

List the recent runs started from this list.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for list collection investigations.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_list_connected_tools`

List Connected Tools · Read-only

List supported tables, CRM, automation and chat tools.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_connection_destinations`

List Connection Destinations · Read-only

List provider-authorized destinations.
Returns a JSON object with data.
Needs identity (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `identity` (`string`, required): Identity for list connection destinations.
- `base_id` (`string`, optional): Base ID for list connection destinations.
- `cursor` (`string`, optional): Cursor for list connection destinations.

Example input: `{"identity":"<identity>"}`

### `specstream_list_connections`

List Connections · Read-only

List this account's connected providers.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_customer_webhooks`

List Customer Webhooks · Read-only

List owned callback destinations without secrets.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_destination_exports`

List Destination Exports · Read-only

List delivery receipts for connected tools.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_destination_syncs`

List Destination Syncs · Read-only

List continuous destination deliveries.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_investigation_functions`

List Investigation Functions · Read-only

List the saved requests in this project that can run once per row.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `project_id` (`string`, optional): Project ID for list investigation functions.

Example input: `{}`

### `specstream_list_project_members`

List Project Members · Read-only

List current project membership.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for list project members.
- `cursor` (`string`, optional): Cursor for list project members.
- `limit` (`integer`, optional): Limit for list project members. Default: `50`.

Example input: `{"id":"<id>"}`

### `specstream_list_projects`

List Projects · Read-only

List currently authorized projects.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for list projects.
- `limit` (`integer`, optional): Limit for list projects. Default: `50`.

Example input: `{}`

### `specstream_list_reconciliations`

List Reconciliations · Read-only

List the recent comparisons that include this list (the latest 20, kept for 30 days).
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for list reconciliations.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_list_request_templates`

List Request Templates · Read-only

List request examples; compile and quote before work.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `record_type` (`string`, optional): Record Type for list request templates.

Example input: `{}`

### `specstream_list_research`

List Research · Read-only

List your research runs, newest first; follow next_after.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): The run id next_after named: the page after it.
- `limit` (`integer`, optional): Limit for list research. Default: `20`.

Example input: `{}`

### `specstream_list_research_rules`

List Research Rules · Read-only

List the automatic research rules on this list.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for list research rules.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_list_search_dialects`

List Search Dialects · Read-only

List the other search and page APIs SpecStream answers in, each with the addresses to paste and a curl call; no key needed.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_search_engines`

List Search Engines · Read-only

List the web search engines a Search API call may ask, each with who runs it and the country it is run from, for a key's engine_rule and a call's engines parameter; no key needed.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_list_spec_finds`

List Spec Finds · Read-only

Read the finds a saved request has run.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for list spec finds.
- `cursor` (`string`, optional): Cursor for list spec finds.
- `limit` (`integer`, optional): Limit for list spec finds. Default: `50`.

Example input: `{"id":"<id>"}`

### `specstream_list_spec_revisions`

List Spec Revisions · Read-only

Read a saved request's revision history.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for list spec revisions.
- `cursor` (`string`, optional): Cursor for list spec revisions.
- `limit` (`integer`, optional): Limit for list spec revisions. Default: `50`.

Example input: `{"id":"<id>"}`

### `specstream_list_specs`

List Specs · Read-only

List the saved requests you can open.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `archived` (`boolean`, optional): Archived for list specs. Default: `false`.
- `cursor` (`string`, optional): Cursor for list specs.
- `limit` (`integer`, optional): Limit for list specs. Default: `50`.
- `project_id` (`string`, optional): Project ID for list specs.

Example input: `{}`

### `specstream_list_web_sources`

List Web Sources · Read-only

List up to 100 saved website sources in a project. No page is read.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_map_site`

Map Site · Read-only

Map a site: its start page's links and its sitemaps' pages, kept to the site. One page read when it gives a link.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `url` (`string`, required): URL for map site.
- `idempotency_key` (`string`, optional): Your id for this call, at most 128 characters. The same id again answers with the same map charge or the same crawl job.
- `include_subdomains` (`boolean`, optional): Include Subdomains for map site. Default: `false`.
- `limit` (`integer`, optional): Limit for map site. Default: `5000`.
- `sitemap` (`string`, optional): Sitemap for map site. One of: include, skip, only. One of: `include`, `skip`, `only`. Default: `include`.

Example input: `{"url":"<url>"}`

### `specstream_parse_from_chips`

Parse From Chips · Read-only

Edit the reading with typed changes, without a model call.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ast` (`object`, required): Ast for parse from chips.
- `operations` (`array`, required): Operations for parse from chips.

Example input: `{"ast":{},"operations":[]}`

### `specstream_parse_query`

Parse Query · Read-only

Read a request in plain words and return an editable reading.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): Q for parse query.
- `expected_source_hash` (`string`, optional): Expected Source Hash for parse query.
- `kind` (`string`, optional): Chosen result kind; the compiler never switches away from it.
- `locale` (`string`, optional): Locale for parse query. Default: `en-US`.
- `named_phrases` (`array`, optional): Phrases the caller knows are one name each (a page's own subject, a product line with commas in it). Each is read as one name wherever it appears in q; a phrase that does not appear in q is ignored.
- `retained` (`object`, optional): An explicit dependency on an authorized, immutable prior research request.

The label is display text only. It never selects records or changes the target.
The server loads the actual Spec; callers cannot provide a replacement base.

Example input: `{"q":"<q>"}`

### `specstream_preview_cloud_list`

Preview Cloud List · Read-only

Read a provider-authorized list before importing.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `connection_id` (`string`, required): Connection ID for preview cloud list.
- `link` (`string`, required): Link for preview cloud list.
- `sheet_id` (`string`, optional): Sheet ID for preview cloud list.

Example input: `{"connection_id":"<connection_id>","link":"<link>"}`

### `specstream_preview_web_source`

Preview Web Source · Read-only

Read the saved page again and preview its rows. Earlier lists stay unchanged. Columns paired by position need fresh examples once the page changes.
Returns a JSON object with data.
Needs source_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `source_id` (`string`, required): Source ID for preview web source.

Example input: `{"source_id":"<source_id>"}`

### `specstream_quote_capacity_work`

Quote Capacity Work · Read-only

Get a quote for a request before anything runs.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ast_hash` (`string`, required): Ast Hash for quote capacity work.
- `minimum_units` (`integer`, optional): Minimum Units for quote capacity work.
- `operation` (`string`, optional): Operation for quote capacity work. One of: find, extend, add_facts, refresh, retained. One of: `find`, `extend`, `add_facts`, `refresh`, `retained`. Default: `find`.
- `parent_find_id` (`string`, optional): Parent Find ID for quote capacity work.
- `project_id` (`string`, optional): Project ID for quote capacity work.
- `requested_units` (`integer`, optional): Requested Units for quote capacity work. Default: `25`.
- `revision_id` (`string`, optional): Revision ID for quote capacity work.
- `selected_entity_ids` (`array`, optional): Selected Entity IDS for quote capacity work.
- `source_import_id` (`string`, optional): Source Import ID for quote capacity work.
- `source_row_ids` (`array`, optional): Source Row IDS for quote capacity work.

Example input: `{"ast_hash":"<ast_hash>"}`

### `specstream_replay_find_decisions`

Replay Find Decisions · Read-only

Replay saved match decisions from their saved sources, with no new source or model calls.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for replay find decisions.

Example input: `{"id":"<id>"}`

### `specstream_research_events`

Research Events · Read-only

A research run's events as a server-sent stream, to its end; Last-Event-ID resumes it.
Needs run_id (identifiers returned by earlier SpecStream calls).
The response is an event stream returned as text: the first lines show the events received before the connection closes.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `run_id` (`string`, required): Run ID for research events.
- `after` (`integer`, optional): Send the events after this event id (as Last-Event-ID does).
- `last_event_id` (`string`, optional): Optional value sent in the Last-Event-Id HTTP header.

Example input: `{"run_id":"<run_id>"}`

### `specstream_teach_web_source`

Teach Web Source · Read-only

Read one public page and learn its records and fields from 3 to 8 example rows. The preview explains how each field was inferred and marks the example rows. Nothing becomes a list until you save it.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `examples` (`array`, required): Examples for teach web source.
- `name` (`string`, required): Name for teach web source.
- `url` (`string`, required): URL for teach web source.
- `exclude` (`array`, optional): Exclude for teach web source.
- `name_column` (`string`, optional): Name Column for teach web source. Default: `name`.
- `project_id` (`string`, optional): Project ID for teach web source.

Example input: `{"examples":[],"name":"<name>","url":"<url>"}`

### `specstream_understand_live`

Understand Live · Read-only

Read an exact editor snapshot without saving or authorizing work.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `document_id` (`string`, required): Document ID for understand live.
- `revision` (`integer`, required): Revision for understand live.
- `text` (`string`, required): Text for understand live.
- `context` (`object`, optional): Workspace hints. They never widen the request. The quote holds the context of the run.
- `expected_source_hash` (`string`, optional): Expected Source Hash for understand live.
- `generation` (`integer`, optional): Generation for understand live. Default: `0`.
- `kind` (`string`, optional): Chosen result kind; shared by fast and refined readings.
- `phase` (`string`, optional): Phase for understand live. One of: fast, refined. One of: `fast`, `refined`. Default: `fast`.

Example input: `{"document_id":"<document_id>","revision":1,"text":"<text>"}`

### `specstream_web_search`

Web Search · Read-only

Search API: answer a question from the live web in seconds, each value with its source.
With no API key, the call answers 402 with a PAYMENT-REQUIRED header (base64 x402 v2 PaymentRequired): pay for this one call with x402's `exact` scheme in USDC on Base, Arbitrum or Polygon, and send the same request again with the payment in PAYMENT-SIGNATURE. The reply then carries PAYMENT-RESPONSE naming the transaction. One search of up to 10 results costs $0.007 ($0.002 with speed=fast, news or images), and a page read costs $0.002 for each page you ask for. The price is taken when the payment settles, before the call runs, so a search with no result or a page with no text is paid. If the call fails on our side after the payment settled, or the settle has no answer yet, send the same request with the same PAYMENT-SIGNATURE again: it is answered, and nothing is taken twice.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): The question or the request in everyday words, in any language. Search operators hold on every result: site:, -site:, -word, -"phrase", "phrase", filetype:, after: and before: (YYYY, YYYY-MM or YYYY-MM-DD). A result that breaks one is left out; when that leaves a page short, the engines' next pages are searched first.
- `country` (`string`, optional): The country the results are for, as a two-letter ISO 3166-1 code (DE). Sent to every search engine that takes a country.
- `engines` (`string`, optional): Narrows the engines this call may ask, never past the key's own engine_rule: engine names and two-letter ISO 3166-1 country codes of their operators, comma-separated. A leading - denies one (-CN asks no engine run from China); names without - are the only ones asked (bing,duckduckgo). The engines are listed at /v1/search-engines.
- `engines_asked` (`boolean`, optional): Name the engines the query was sent to in the reply (engines_asked). Default: `false`.
- `idempotency_key` (`string`, optional): Your id for this call, at most 128 characters. A call is charged once per id, and the same id again is refused with 409. Without one, SpecStream gives the call an id and returns it as request_id.
- `language` (`string`, optional): The language of the results, as a two-letter ISO 639-1 code (de). Sent to every search engine that takes a language.
- `news` (`boolean`, optional): News items for the query instead of web results: newest first, each with its publisher (source), its day (published) and its time (published_at). A fast search at the same price; not with passages. Default: `false`.
- `num` (`integer`, optional): The number of results (sources) to return, 1 to 100, or 1 to 200 with images=true. Without num or page, the reply holds every result the answer read. When the first search finds fewer, SpecStream searches deeper (more engines and their next result pages) before it answers. A call of up to 10 results is $0.005; each started block of 10 more results adds $0.006 (100 results are $0.059). On the fast tier (speed=fast) a call of up to 10 results is $0.001 and each started block of 10 more adds $0.0012 (100 results are $0.0118). An image search (images=true) always runs the fast tier, so 200 images are $0.0238. A call is charged for the results it returns; a call that returns none costs nothing.
- `page` (`integer`, optional): The page of results to return, 1 to 10, num results each (10 without num). Page 2 continues the same list as page 1. The answer and the rows are on page 1 only. next_page in the reply names the next page, or is null when the engines found no more.
- `pages_wait_ms` (`integer`, optional): With passages=true: how long, in milliseconds, the call waits for the pages' text. The first five sources wait for theirs; the rest keep what is in by then, and pages_pending names the sources still being read. Absent: the call waits for every source's text, as before.
- `passages` (`boolean`, optional): Also return the text of the top pages closest to the request, each with its page, so a caller sees the evidence without reading a page. Default: `false`.
- `safe` (`string`, optional): The safe search level: off, moderate or strict. Sent to every search engine that takes one; without it each engine uses its own default. One of: off, moderate, strict. One of: `off`, `moderate`, `strict`.
- `speed` (`string`, optional): fast gives the search results in about a second: the first engines' results with their snippets, each with the time it was read (read_at), with no answer, rows or passages; num and page past the first 10 results come from the engines' own later pages. default also reads the results for the answer and its rows. A fast call of up to 10 results is $0.001 ($1 per 1,000), a default call $0.005 ($5 per 1,000). One of: default, fast. One of: `default`, `fast`. Default: `default`.
- `x_end_user_cap_cents` (`integer`, optional): The most the end user named in X-End-User-Id may spend this UTC month, in cents. A call past it is refused with 402 and details.reason end_user_cap. Send it on each call; SpecStream stores no policy.
- `x_end_user_id` (`string`, optional): Your own label for the end user this call is for. SpecStream keeps only its SHA-256 hex, and GET /v1/account/usage?group=end_user sums your calls by it.

Example input: `{"q":"<q>"}`

### `specstream_web_search_answer`

Web Search Answer · Read-only

Open one kept web search answer as it was given.
Returns a JSON object with data.
Needs answer_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `answer_id` (`string`, required): Answer ID for web search answer.

Example input: `{"answer_id":"<answer_id>"}`

### `specstream_web_search_answers`

Web Search Answers · Read-only

List your web search history: each question, its answer and its sources, newest first.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for web search answers.
- `limit` (`integer`, optional): Limit for web search answers. Default: `20`.
- `q` (`string`, optional): Searches whose words contain this text, in any case.

Example input: `{}`

### `specstream_web_search_batch`

Web Search Batch · Read-only

Search API for many questions in one call, each answered and billed as one web search.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `questions` (`array`, required): Questions for web search batch.
- `x_end_user_cap_cents` (`integer`, optional): The most the end user named in X-End-User-Id may spend this UTC month, in cents. A call past it is refused with 402 and details.reason end_user_cap. Send it on each call; SpecStream stores no policy.
- `x_end_user_id` (`string`, optional): Your own label for the end user this call is for. SpecStream keeps only its SHA-256 hex, and GET /v1/account/usage?group=end_user sums your calls by it.

Example input: `{"questions":[]}`

### `specstream_web_search_post`

Web Search Post · Read-only

Search API with the question in a JSON body: the GET's parameters, same names, defaults, price and reply, so no address and no log of one holds the question's words. A parameter sent in both the address and the body is refused.
With no API key, the call answers 402 with a PAYMENT-REQUIRED header (base64 x402 v2 PaymentRequired): pay for this one call with x402's `exact` scheme in USDC on Base, Arbitrum or Polygon, and send the same request again with the payment in PAYMENT-SIGNATURE. The reply then carries PAYMENT-RESPONSE naming the transaction. One search of up to 10 results costs $0.007 ($0.002 with speed=fast, news or images), and a page read costs $0.002 for each page you ask for. The price is taken when the payment settles, before the call runs, so a search with no result or a page with no text is paid. If the call fails on our side after the payment settled, or the settle has no answer yet, send the same request with the same PAYMENT-SIGNATURE again: it is answered, and nothing is taken twice.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): Q for web search post.
- `country` (`string`, optional): Country for web search post.
- `engines` (`string`, optional): Engines for web search post.
- `engines_asked` (`boolean`, optional): Engines Asked for web search post. Default: `false`.
- `idempotency_key` (`string`, optional): Your id for this call, at most 128 characters. A call is charged once per id, and the same id again is refused with 409. Without one, SpecStream gives the call an id and returns it as request_id.
- `image_aspect` (`string`, optional): Image Aspect for web search post.
- `image_color` (`string`, optional): Image Color for web search post.
- `image_size` (`string`, optional): Image Size for web search post.
- `image_type` (`string`, optional): Image Type for web search post.
- `images` (`boolean`, optional): Images for web search post. Default: `false`.
- `language` (`string`, optional): Language for web search post.
- `max_height` (`integer`, optional): Max Height for web search post.
- `max_width` (`integer`, optional): Max Width for web search post.
- `min_height` (`integer`, optional): Min Height for web search post.
- `min_width` (`integer`, optional): Min Width for web search post.
- `news` (`boolean`, optional): News for web search post. Default: `false`.
- `num` (`integer`, optional): Num for web search post.
- `page` (`integer`, optional): Page for web search post.
- `pages_wait_ms` (`integer`, optional): Pages Wait Ms for web search post.
- `passages` (`boolean`, optional): Passages for web search post. Default: `false`.
- `safe` (`string`, optional): Safe for web search post.
- `speed` (`string`, optional): Speed for web search post. Default: `default`.
- `x_end_user_cap_cents` (`integer`, optional): The most the end user named in X-End-User-Id may spend this UTC month, in cents. A call past it is refused with 402 and details.reason end_user_cap. Send it on each call; SpecStream stores no policy.
- `x_end_user_id` (`string`, optional): Your own label for the end user this call is for. SpecStream keeps only its SHA-256 hex, and GET /v1/account/usage?group=end_user sums your calls by it.

Example input: `{"q":"<q>"}`

### `specstream_workspace_activity`

Workspace Activity · Read-only

Read activity through current account and project permissions.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for workspace activity.
- `limit` (`integer`, optional): Limit for workspace activity. Default: `50`.
- `project_id` (`string`, optional): Project ID for workspace activity.

Example input: `{}`

### `specstream_workspace_collection_changes`

Workspace Collection Changes · Read-only

Read source-backed changes against immutable prior results.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace collection changes.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_export_audit_events`

Workspace Export Audit Events · Read-only

Export an authorized activity page; follow X-Next-Cursor.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `actor_account_id` (`string`, optional): Actor Account ID for workspace export audit events.
- `cursor` (`string`, optional): Cursor for workspace export audit events.
- `kind` (`string`, optional): Kind for workspace export audit events. One of: Sign-in, Access, Spending, Exports, Automation, Lists. One of: `Sign-in`, `Access`, `Spending`, `Exports`, `Automation`, `Lists`.
- `limit` (`integer`, optional): Limit for workspace export audit events. Default: `100`.
- `project_id` (`string`, optional): Project ID for workspace export audit events.

Example input: `{}`

### `specstream_workspace_export_collection`

Workspace Export Collection · Read-only

Export a full collection with current scoped access and suppression.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace export collection.
- `format` (`string`, optional): Format for workspace export collection. One of: csv, xlsx, json, html, kml, geojson. One of: `csv`, `xlsx`, `json`, `html`, `kml`, `geojson`. Default: `csv`.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_filter_collection`

Workspace Filter Collection · Read-only

Filter a list's rows by IDs, sources or a map area.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace filter collection.
- `bbox` (`array`, optional): Bbox for workspace filter collection.
- `record_ids` (`array`, optional): Record IDS for workspace filter collection.
- `source_collection_ids` (`array`, optional): Source Collection IDS for workspace filter collection.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_launch_native_provider`

Workspace Launch Native Provider · Read-only

Launch the browser-bound provider handoff.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ticket` (`string`, required): Ticket for workspace launch native provider.

Example input: `{"ticket":"<ticket>"}`

### `specstream_workspace_list_audit_events`

Workspace List Audit Events · Read-only

Read immutable twelve-month activity with current owner permissions.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `actor_account_id` (`string`, optional): Actor Account ID for workspace list audit events.
- `cursor` (`string`, optional): Cursor for workspace list audit events.
- `kind` (`string`, optional): Kind for workspace list audit events. One of: Sign-in, Access, Spending, Exports, Automation, Lists. One of: `Sign-in`, `Access`, `Spending`, `Exports`, `Automation`, `Lists`.
- `limit` (`integer`, optional): Limit for workspace list audit events. Default: `50`.
- `project_id` (`string`, optional): Project ID for workspace list audit events.

Example input: `{}`

### `specstream_workspace_list_collections`

Workspace List Collections · Read-only

Read the lists you can open, from your finds and imports.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): After for workspace list collections.
- `limit` (`integer`, optional): Limit for workspace list collections. Default: `50`.
- `project_id` (`string`, optional): Project ID for workspace list collections.
- `sort` (`string`, optional): Sort for workspace list collections. One of: id, recent. One of: `id`, `recent`. Default: `id`.

Example input: `{}`

### `specstream_workspace_list_notes`

Workspace List Notes · Read-only

Read authorized project or collection-record notes.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, optional): Collection ID for workspace list notes.
- `project_id` (`string`, optional): Project ID for workspace list notes.
- `record_id` (`string`, optional): Record ID for workspace list notes.

Example input: `{}`

### `specstream_workspace_list_notifications`

Workspace List Notifications · Read-only

Read your inbox with current resource authorization.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Cursor for workspace list notifications.
- `limit` (`integer`, optional): Limit for workspace list notifications. Default: `50`.
- `tab` (`string`, optional): Tab for workspace list notifications. One of: all, unread, archived. One of: `all`, `unread`, `archived`. Default: `all`.

Example input: `{}`

### `specstream_workspace_list_passkeys`

Workspace List Passkeys · Read-only

List the current account's registered passkeys.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_list_push_devices`

Workspace List Push Devices · Read-only

List your registered notification devices.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_list_sso_domains`

Workspace List SSO Domains · Read-only

List account-owned verified SSO domains.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_list_stars`

Workspace List Stars · Read-only

List starred work that remains currently authorized.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_list_views`

Workspace List Views · Read-only

List your saved views of a list.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace list views.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_list_your_fields`

Workspace List Your Fields · Read-only

Read the fields you added to a collection and their values.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace list your fields.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_placement_allowance`

Workspace Placement Allowance · Read-only

Read this billing account's UTC monthly free placement allowance.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_placement_status`

Workspace Placement Status · Read-only

Read placement progress and per-row geocoding outcomes.
Returns a JSON object with data.
Needs import_id, placement_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `import_id` (`string`, required): Import ID for workspace placement status.
- `placement_id` (`string`, required): Placement ID for workspace placement status.

Example input: `{"import_id":"<import_id>","placement_id":"<placement_id>"}`

### `specstream_workspace_push_capabilities`

Workspace Push Capabilities · Read-only

Read configured notification transports, never credentials.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_quote_placement`

Workspace Quote Placement · Read-only

Quote the maximum cost of placing imported addresses on the map.
Returns a JSON object with data.
Needs import_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `import_id` (`string`, required): Import ID for workspace quote placement.

Example input: `{"import_id":"<import_id>"}`

### `specstream_workspace_read_collection`

Workspace Read Collection · Read-only

Read one list, without rows removed by an opt-out.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace read collection.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_read_photo`

Workspace Read Photo · Read-only

Read the account-owned photo version and private URL.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_read_scim_mappings`

Workspace Read SCIM Mappings · Read-only

Read provisioning mappings, provider groups and token state.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace read scim mappings.

Example input: `{"domain_id":"<domain_id>"}`

### `specstream_workspace_relationships`

Workspace Relationships · Read-only

Group accessible records by actual shared column values.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace relationships.
- `field_id` (`string`, optional): Field ID for workspace relationships.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_research_input`

Workspace Research Input · Read-only

Review each imported row: which business it matched and why, other possible matches, and its result.
Returns a JSON object with data.
Needs find_id (identifiers returned by earlier SpecStream calls).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `find_id` (`string`, required): Find ID for workspace research input.

Example input: `{"find_id":"<find_id>"}`

### `specstream_workspace_search`

Workspace Search · Read-only

Search the requests, lists and sources you can open.
Returns a JSON object with data.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `q` (`string`, required): Q for workspace search.
- `cursor` (`string`, optional): Cursor for workspace search.
- `limit` (`integer`, optional): Limit for workspace search. Default: `30`.

Example input: `{"q":"<q>"}`

### `specstream_apply_reconciliation`

Apply Reconciliation · Write

Fill empty cells in a new list from the pairs you chose. The original lists stay unchanged.
Returns a JSON object with data.
Needs comparison_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `choices` (`array`, required): Choices for apply reconciliation.
- `comparison_hash` (`string`, required): Comparison Hash for apply reconciliation.
- `comparison_id` (`string`, required): Comparison ID for apply reconciliation.
- `import_id` (`string`, required): Import ID for apply reconciliation.
- `name` (`string`, required): Name for apply reconciliation.

Example input: `{"choices":[],"comparison_hash":"<comparison_hash>","comparison_id":"<comparison_id>","import_id":"<import_id>","name":"<name>"}`

### `specstream_approve_investigation_batch`

Approve Investigation Batch · Write

Approve the quote's exact maximum. The searches then run as ordinary finds, within that maximum.
Returns a JSON object with data.
Needs batch_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `batch_id` (`string`, required): Batch ID for approve investigation batch.
- `confirm_max_cents` (`integer`, required): Confirm Max Cents for approve investigation batch.
- `quote_hash` (`string`, required): Quote Hash for approve investigation batch.

Example input: `{"batch_id":"<batch_id>","confirm_max_cents":1,"quote_hash":"<quote_hash>"}`

### `specstream_approve_research_rule`

Approve Research Rule · Write

Approve the rule's total spending limit. Rows that qualify later then run without asking again, until the limit or the expiry is reached.
Returns a JSON object with data.
Needs rule_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `confirm_total_cents` (`integer`, required): Confirm Total Cents for approve research rule.
- `review_hash` (`string`, required): Review Hash for approve research rule.
- `rule_id` (`string`, required): Rule ID for approve research rule.

Example input: `{"confirm_total_cents":1,"review_hash":"<review_hash>","rule_id":"<rule_id>"}`

### `specstream_archive_spec`

Archive Spec · Write

Archive or restore a saved request.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_revision_id` (`string`, required): Expected Revision ID for archive spec.
- `id` (`string`, required): ID for archive spec.
- `archived` (`boolean`, optional): Archived for archive spec. Default: `true`.

Example input: `{"expected_revision_id":"<expected_revision_id>","id":"<id>"}`

### `specstream_authorize_capacity_work`

Authorize Capacity Work · Write

Approve the exact quote to start one find and reserve its maximum from your balance. Needs the specs:write, finds:write and finds:spend scopes.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `confirm_max_cents` (`integer`, required): Confirm Max Cents for authorize capacity work.
- `idempotency_key` (`string`, required): Idempotency Key for authorize capacity work.
- `quote_hash` (`string`, required): Quote Hash for authorize capacity work.
- `quote_id` (`string`, required): Quote ID for authorize capacity work.
- `authority_id` (`string`, optional): Authority ID for authorize capacity work.

Example input: `{"confirm_max_cents":1,"idempotency_key":"<idempotency_key>","quote_hash":"<quote_hash>","quote_id":"<quote_id>"}`

### `specstream_begin_connection_authorization`

Begin Connection Authorization · Write

Begin account-bound provider authorization.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `provider` (`string`, required): Provider for begin connection authorization.

Example input: `{"provider":"<provider>"}`

### `specstream_browser_action`

Browser Action · Write

Run one step in the browser: a Playwright MCP tool by its name and arguments (navigate, snapshot, click, type, take a screenshot and the others), or browser_read for the page's text as Markdown.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for browser action.
- `tool` (`string`, required): A Playwright MCP tool name (browser_navigate, browser_snapshot, browser_click, browser_type, browser_take_screenshot and the others), or browser_read for the page's text.
- `arguments` (`object`, optional): The tool's arguments, as Playwright MCP names them. browser_read takes format: markdown (default) or text.

Example input: `{"id":"<id>","tool":"<tool>"}`

### `specstream_buy_agent_list`

Buy Agent List · Write

Buy a list from the live web, priced per result.
Without a payment this answers 402 with the most the run can cost. Pay with MPP or x402 and send the same request again to start the run.
Returns a JSON object with purchase_id, token, status_url, read_with, paid, authorized, and more.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `request` (`string`, required): The list you need, in plain words.
- `count` (`integer`, optional): How many matches to look for.

Example input: `{"request":"<request>"}`

### `specstream_capture_capacity_referral`

Capture Capacity Referral · Write

Remember the first referral visit for sign-up within thirty days.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `ref` (`string`, required): Ref for capture capacity referral.

Example input: `{"ref":"<ref>"}`

### `specstream_change_capacity_membership`

Change Capacity Membership · Write

Change, cancel, or resume the membership for its next renewal.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `action` (`string`, required): Action for change capacity membership. One of: change, cancel, resume. One of: `change`, `cancel`, `resume`.
- `idempotency_key` (`string`, required): Idempotency Key for change capacity membership.
- `plan_id` (`string`, optional): Plan ID for change capacity membership.

Example input: `{"action":"change","idempotency_key":"<idempotency_key>"}`

### `specstream_claim_free_capacity`

Claim Free Capacity · Write

Claim the one-time free balance for a verified eligible account.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs: none.

Example input: `{}`

### `specstream_create_account_key`

Create Account Key · Write

Create ss_live_ or ss_test_ key.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `engine_rule` (`object`, optional): The engines a key's searches may ask (listed at /v1/search-engines): each item an
engine's name or the two-letter ISO 3166-1 code of the country its operator runs it from.
An engine is asked when ``allow`` is empty or names it or its country, and ``deny``
names neither.
- `eu_egress_only` (`boolean`, optional): Eu Egress Only for create account key. Default: `false`.
- `expires_in_days` (`integer`, optional): Days until the key stops working. Without it, or with null, the key works until it is revoked.
- `name` (`string`, optional): Name for create account key. Default: `key`.
- `project_id` (`string`, optional): Project ID for create account key.
- `scopes` (`array`, optional): Scopes for create account key.
- `spend_cap_cents` (`integer`, optional): The most the key spends in total, in cents, counted over every call since the key was made. The account's balance and limits apply as well. Default: `0`.
- `zero_retention` (`boolean`, optional): Zero Retention for create account key. Default: `false`.

Example input: `{}`

### `specstream_create_browser_session`

Create Browser Session · Write

Open a browser: a real Chromium you drive over the Chrome DevTools Protocol with Playwright, Puppeteer, browser-use or Stagehand, or step by step through actions_url. Charged per second it runs and per action.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `allowed_domains` (`array`, optional): Hosts the browser may open, such as example.com or *.example.com. Empty allows every public host. A page outside them is refused with blocked.
- `block_media` (`boolean`, optional): true skips images, video, audio and fonts: pages load faster and the session reads fewer bytes. Default: `false`.
- `headless` (`boolean`, optional): false runs the browser with a screen, for sites that act differently without one. It uses more memory. Default: `true`.
- `idempotency_key` (`string`, optional): Your id for this call, at most 128 characters. The same id again answers with the same session and opens no second one.
- `idle_timeout_s` (`integer`, optional): The session ends after this many seconds with no client connected and no action. With keep_alive false, a disconnect ends it at once. Default: `60`.
- `keep_alive` (`boolean`, optional): true keeps the session when the client disconnects, so a new client can connect, until idle_timeout_s passes. false ends it at the disconnect. Default: `false`.
- `metadata` (`object`, optional): Your own labels for the session, returned as sent.
- `timeout_s` (`integer`, optional): The most seconds the session runs. It ends then, connected or not. Default: `300`.
- `viewport` (`object`, optional): Viewport for create browser session.

Example input: `{}`

### `specstream_create_capacity_authority`

Create Capacity Authority · Write

Approve a spending limit for one agent or member in a project.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `actor_key` (`string`, required): Actor Key for create capacity authority.
- `operations` (`array`, required): Operations for create capacity authority.
- `per_run_cents` (`integer`, required): Per Run Cents for create capacity authority.
- `period_cents` (`integer`, required): Period Cents for create capacity authority.
- `valid_until` (`string`, required): Valid Until for create capacity authority.
- `project_id` (`string`, optional): Project ID for create capacity authority.
- `quote_id` (`string`, optional): Quote ID for create capacity authority.
- `scope_hash` (`string`, optional): Scope Hash for create capacity authority.

Example input: `{"actor_key":"<actor_key>","operations":[],"per_run_cents":1,"period_cents":1,"valid_until":"<valid_until>"}`

### `specstream_create_capacity_funding`

Create Capacity Funding · Write

Prepare owner-authorized balance or membership checkout without starting work.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `idempotency_key` (`string`, required): Idempotency Key for create capacity funding.
- `kind` (`string`, required): Kind for create capacity funding. One of: topup, subscription, search_pack. One of: `topup`, `subscription`, `search_pack`.
- `amount_cents` (`integer`, optional): Amount Cents for create capacity funding.
- `currency` (`string`, optional): Currency for create capacity funding. One of: usd, eur. One of: `usd`, `eur`.
- `intended_quote_id` (`string`, optional): Intended Quote ID for create capacity funding.
- `plan_id` (`string`, optional): Plan ID for create capacity funding.
- `search_pack_id` (`string`, optional): Search Pack ID for create capacity funding.

Example input: `{"idempotency_key":"<idempotency_key>","kind":"topup"}`

### `specstream_create_capacity_refund_claim`

Create Capacity Refund Claim · Write

Report a wrong result for review against its sources.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `idempotency_key` (`string`, required): Idempotency Key for create capacity refund claim.
- `reason` (`string`, required): Reason for create capacity refund claim.
- `reservation_id` (`string`, required): Reservation ID for create capacity refund claim.
- `unit_ids` (`array`, required): Unit IDS for create capacity refund claim.

Example input: `{"idempotency_key":"<idempotency_key>","reason":"<reason>","reservation_id":"<reservation_id>","unit_ids":[]}`

### `specstream_create_capacity_schedule`

Create Capacity Schedule · Write

Approve future runs of an exact request within set spending limits.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `cadence_seconds` (`integer`, required): Cadence Seconds for create capacity schedule.
- `first_due_at` (`string`, required): First Due At for create capacity schedule.
- `idempotency_key` (`string`, required): Idempotency Key for create capacity schedule.
- `per_run_cents` (`integer`, required): Per Run Cents for create capacity schedule.
- `period_cents` (`integer`, required): Period Cents for create capacity schedule.
- `scope` (`object`, required): Scope for create capacity schedule.
- `valid_until` (`string`, required): Valid Until for create capacity schedule.
- `calendar_schedule` (`object`, optional): Calendar Schedule for create capacity schedule.

Example input: `{"cadence_seconds":1,"first_due_at":"<first_due_at>","idempotency_key":"<idempotency_key>","per_run_cents":1,"period_cents":1,"scope":{},"valid_until":"<valid_until>"}`

### `specstream_create_customer_webhook`

Create Customer Webhook · Write

Register find.completed webhook.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `url` (`string`, required): URL for create customer webhook.
- `event_name` (`string`, optional): Event Name for create customer webhook. One of: find.completed, monitor.found, collection.records_added, auto_topup.paid, auto_topup.failed, balance.low, research.completed, research.failed, research.cancelled. One of: `find.completed`, `monitor.found`, `collection.records_added`, `auto_topup.paid`, `auto_topup.failed`, `balance.low`, `research.completed`, `research.failed`, `research.cancelled`. Default: `find.completed`.
- `events` (`array`, optional): Events for create customer webhook.

Example input: `{"url":"<url>"}`

### `specstream_create_destination_sync`

Create Destination Sync · Write

Keep sending new and changed records to this destination.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collectionId` (`string`, required): Collectionid for create destination sync.
- `connection_id` (`string`, required): Connection ID for create destination sync.
- `destination_sha256` (`string`, required): Destination Sha256 for create destination sync.
- `keepSending` (`boolean`, optional): Keepsending for create destination sync. Default: `true`.

Example input: `{"collectionId":"<collectionId>","connection_id":"<connection_id>","destination_sha256":"<destination_sha256>"}`

### `specstream_create_investigation_function`

Create Investigation Function · Write

Make a saved request run once per row by turning some of its values into typed inputs. Nothing runs and nothing is charged.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name for create investigation function.
- `parameters` (`array`, required): Parameters for create investigation function.
- `revision_id` (`string`, required): Revision ID for create investigation function.
- `spec_id` (`string`, required): Spec ID for create investigation function.

Example input: `{"name":"<name>","parameters":[],"revision_id":"<revision_id>","spec_id":"<spec_id>"}`

### `specstream_create_project`

Create Project · Write

Create a project with an explicit bounded authorization ceiling.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name for create project.
- `monthly_spend_cap_cents` (`integer`, optional): Monthly Spend Cap Cents for create project. Default: `0`.

Example input: `{"name":"<name>"}`

### `specstream_create_research`

Create Research · Write

Start a research report: the run plans the question, searches the live web, reads pages and writes a report in which each cited sentence links to the page it was read from. Answers at once with the run's id; charged once, at delivery, at the depth's price.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `input` (`string`, required): Input for create research.
- `after_date` (`string`, optional): After Date for create research.
- `depth` (`string`, optional): Depth for create research. One of: quick, standard, deep. One of: `quick`, `standard`, `deep`. Default: `standard`.
- `exclude_domains` (`array`, optional): Exclude Domains for create research.
- `idempotency_key` (`string`, optional): Your id for this run, at most 128 characters. The same id with the same body answers with the same run; with another body it is refused with 409.
- `include_domains` (`array`, optional): Include Domains for create research.
- `language` (`string`, optional): Language for create research.
- `length` (`string`, optional): Length for create research. One of: short, standard, long. One of: `short`, `standard`, `long`. Default: `standard`.
- `metadata` (`object`, optional): Metadata for create research.
- `output` (`object`, optional): Output for create research.
- `webhook` (`object`, optional): Webhook for create research.

Example input: `{"input":"<input>"}`

### `specstream_create_spec`

Create Spec · Write

Save a request as its first revision.
Returns a JSON object with data.
Provide exactly one of: ast, source_find_id.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name for create spec.
- `ast` (`object`, optional): Ast for create spec.
- `context` (`string`, optional): Context for create spec. Default: ``.
- `project_id` (`string`, optional): Project ID for create spec.
- `q` (`string`, optional): The request in the person's own words. Lists run from this version are named from it. Without it, the lists are named from the reading.
- `source_find_id` (`string`, optional): Source Find ID for create spec.

Example input: `{"name":"<name>"}`

### `specstream_create_spec_link`

Create Spec Link · Write

Share a saved request revision as a link, without its results or permission to spend.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_ast_hash` (`string`, required): Expected Ast Hash for create spec link.
- `id` (`string`, required): ID for create spec link.
- `revision_id` (`string`, required): Revision ID for create spec link.
- `expires_seconds` (`integer`, optional): Expires Seconds for create spec link. Default: `2592000`.
- `include_context` (`boolean`, optional): Include Context for create spec link. Default: `false`.

Example input: `{"expected_ast_hash":"<expected_ast_hash>","id":"<id>","revision_id":"<revision_id>"}`

### `specstream_export_find`

Export Find · Write

One-use signed download.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `format` (`string`, required): Owned Delivery export formats, including maps from supplied coordinates. One of: xlsx, csv, json, html, kml, geojson. One of: `xlsx`, `csv`, `json`, `html`, `kml`, `geojson`.
- `id` (`string`, required): ID for export find.

Example input: `{"format":"xlsx","id":"<id>"}`

### `specstream_import_cloud_list`

Import Cloud List · Write

Import supplied records from a connected table or workbook.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `connection_id` (`string`, required): Connection ID for import cloud list.
- `import_id` (`string`, required): Import ID for import cloud list.
- `link` (`string`, required): Link for import cloud list.
- `name` (`string`, required): Name for import cloud list.
- `identity_column` (`string`, optional): Identity Column for import cloud list.
- `keepUpdated` (`boolean`, optional): Keepupdated for import cloud list. Default: `true`.
- `mapping` (`object`, optional): Mapping for import cloud list.
- `project_id` (`string`, optional): Project ID for import cloud list.
- `sheet_id` (`string`, optional): Sheet ID for import cloud list.
- `source_sha256` (`string`, optional): Source Sha256 for import cloud list.

Example input: `{"connection_id":"<connection_id>","import_id":"<import_id>","link":"<link>","name":"<name>"}`

### `specstream_import_file_command`

Import File Command · Write

Import supplied CSV or XLSX values without starting research.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `format` (`string`, required): Format for import file command.
- `import_id` (`string`, required): Import ID for import file command.
- `name` (`string`, required): Name for import file command.
- `content` (`string`, optional): Content for import file command.
- `content_base64` (`string`, optional): Content Base64 for import file command.
- `mapping` (`object`, optional): Mapping for import file command.
- `project_id` (`string`, optional): Project ID for import file command.
- `sheet` (`string`, optional): Sheet for import file command.

Example input: `{"format":"<format>","import_id":"<import_id>","name":"<name>"}`

### `specstream_open_support_ticket`

Open Support Ticket · Write

Correction or report ticket.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `kind` (`string`, required): Kind for open support ticket. One of: correction, report, opt_out, export, delete. One of: `correction`, `report`, `opt_out`, `export`, `delete`.
- `body` (`string`, optional): Body for open support ticket. Default: ``.
- `entity_id` (`string`, optional): Entity ID for open support ticket.
- `find_id` (`string`, optional): Find ID for open support ticket.
- `reason` (`string`, optional): Reason for open support ticket.

Example input: `{"kind":"correction"}`

### `specstream_opt_out`

Opt Out · Write

Public suppression opt-out.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `email` (`string`, optional): Email for opt out.
- `employer` (`string`, optional): Employer for opt out.
- `name` (`string`, optional): Name for opt out.
- `phone` (`string`, optional): Phone for opt out.
- `profile_url` (`string`, optional): Profile URL for opt out.
- `website` (`string`, optional): Website for opt out.

Example input: `{}`

### `specstream_prepare_investigation_batch`

Prepare Investigation Batch · Write

Bind the selected rows to the inputs and return one quote: the distinct searches, rows with missing inputs, searches reused from earlier work, and one maximum. Nothing runs yet.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `bindings` (`object`, required): Bindings for prepare investigation batch.
- `collection_id` (`string`, required): Collection ID for prepare investigation batch.
- `function_id` (`string`, required): Function ID for prepare investigation batch.
- `record_ids` (`array`, required): Record IDS for prepare investigation batch.
- `request_id` (`string`, required): Request ID for prepare investigation batch.
- `reuse_completed` (`boolean`, optional): Reuse Completed for prepare investigation batch. Default: `true`.

Example input: `{"bindings":{},"collection_id":"<collection_id>","function_id":"<function_id>","record_ids":[],"request_id":"<request_id>"}`

### `specstream_prepare_reconciliation`

Prepare Reconciliation · Write

Compare two of your lists with exact or close-spelling rules and conflict checks, and propose which rows correspond.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `fill` (`array`, required): Fill for prepare reconciliation.
- `groups` (`array`, required): Groups for prepare reconciliation.
- `left` (`object`, required): Left for prepare reconciliation.
- `name` (`string`, required): Name for prepare reconciliation.
- `request_id` (`string`, required): Request ID for prepare reconciliation.
- `right` (`object`, required): Right for prepare reconciliation.
- `assignment` (`string`, optional): Assignment for prepare reconciliation. One of: one_to_one, many_to_one. One of: `one_to_one`, `many_to_one`. Default: `one_to_one`.
- `include_private_fields` (`boolean`, optional): Include Private Fields for prepare reconciliation. Default: `false`.
- `project_id` (`string`, optional): Project ID for prepare reconciliation.
- `vetoes` (`array`, optional): Vetoes for prepare reconciliation.

Example input: `{"fill":[],"groups":[],"left":{},"name":"<name>","request_id":"<request_id>","right":{}}`

### `specstream_prepare_research_rule`

Prepare Research Rule · Write

Prepare a rule that researches rows when they first become ready or when their inputs change, with a per-run and a total spending limit. Nothing runs until it's approved.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `bindings` (`object`, required): Bindings for prepare research rule.
- `collection_id` (`string`, required): Collection ID for prepare research rule.
- `function_id` (`string`, required): Function ID for prepare research rule.
- `name` (`string`, required): Name for prepare research rule.
- `per_batch_cents` (`integer`, required): Per Batch Cents for prepare research rule.
- `request_id` (`string`, required): Request ID for prepare research rule.
- `total_cents` (`integer`, required): Total Cents for prepare research rule.
- `valid_until` (`string`, required): Valid Until for prepare research rule.
- `batch_size` (`integer`, optional): Batch Size for prepare research rule. Default: `20`.
- `conditions` (`array`, optional): Conditions for prepare research rule.
- `include_existing` (`boolean`, optional): Include Existing for prepare research rule. Default: `false`.
- `reuse_completed` (`boolean`, optional): Reuse Completed for prepare research rule. Default: `true`.
- `trigger` (`string`, optional): Trigger for prepare research rule. One of: first_ready, input_changed. One of: `first_ready`, `input_changed`. Default: `first_ready`.

Example input: `{"bindings":{},"collection_id":"<collection_id>","function_id":"<function_id>","name":"<name>","per_batch_cents":1,"request_id":"<request_id>","total_cents":1,"valid_until":"<valid_until>"}`

### `specstream_put_project_member`

Put Project Member · Write

Set a current project member role and bounded spending ceiling.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `account_id` (`string`, required): Account ID for put project member.
- `id` (`string`, required): ID for put project member.
- `role` (`string`, required): Role for put project member. One of: editor, viewer. One of: `editor`, `viewer`.
- `monthly_spend_cap_cents` (`integer`, optional): Monthly Spend Cap Cents for put project member. Default: `0`.

Example input: `{"account_id":"<account_id>","id":"<id>","role":"editor"}`

### `specstream_reconcile_destination_export`

Reconcile Destination Export · Write

Check an uncertain delivery without sending it again.
Returns a JSON object with data.
Needs identity, export_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `export_id` (`string`, required): Export ID for reconcile destination export.
- `identity` (`string`, required): Identity for reconcile destination export.

Example input: `{"export_id":"<export_id>","identity":"<identity>"}`

### `specstream_reconcile_find_commerce`

Reconcile Find Commerce · Write

Recover the existing purchase without rerunning work.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for reconcile find commerce.

Example input: `{"id":"<id>"}`

### `specstream_release_browser_session`

Release Browser Session · Write

Close a browser session and settle its charge. A repeat call answers the same.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for release browser session.

Example input: `{"id":"<id>"}`

### `specstream_request_capacity_custom_plan`

Request Capacity Custom Plan · Write

Request an Enterprise conversation without granting an entitlement.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `brief` (`string`, required): Brief for request capacity custom plan.

Example input: `{"brief":"<brief>"}`

### `specstream_restore_capacity_store_purchases`

Restore Capacity Store Purchases · Write

Recover verified purchases without duplicate balance grants.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `purchases` (`array`, required): Purchases for restore capacity store purchases.

Example input: `{"purchases":[]}`

### `specstream_revise_spec`

Revise Spec · Write

Create an immutable revision only if the expected head still matches.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `ast` (`object`, required): Ast for revise spec.
- `expected_revision_id` (`string`, required): Expected Revision ID for revise spec.
- `id` (`string`, required): ID for revise spec.
- `context` (`string`, optional): Context for revise spec.
- `name` (`string`, optional): Name for revise spec.
- `parent_revision_id` (`string`, optional): The revision of the same request that the new version branches from. Without it, the new version branches from the current head. expected_revision_id still protects the current head against concurrent edits.
- `q` (`string`, optional): The revised request in the person's own words. Lists run from this version are named from it. Without it, the lists are named from the reading.

Example input: `{"ast":{},"expected_revision_id":"<expected_revision_id>","id":"<id>"}`

### `specstream_run_capacity_schedule_now`

Run Capacity Schedule Now · Write

Run active recurring work once now, inside its own limits, without moving its schedule.
Returns a JSON object with data.
Needs schedule_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `schedule_id` (`string`, required): Schedule ID for run capacity schedule now.

Example input: `{"schedule_id":"<schedule_id>"}`

### `specstream_save_connection_destination`

Save Connection Destination · Write

Save a provider-verified destination and field mapping.
Returns a JSON object with data.
Needs identity (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `identity` (`string`, required): Identity for save connection destination.
- `selection` (`object`, required): Selection for save connection destination.

Example input: `{"identity":"<identity>","selection":{}}`

### `specstream_save_web_source_list`

Save Web Source List · Write

Save a reviewed preview as an imported list, using its preview hash. Saving does not research or spend; quote and approve research on the list afterward.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `import_id` (`string`, required): Import ID for save web source list.
- `name` (`string`, required): Name for save web source list.
- `preview_hash` (`string`, required): Preview Hash for save web source list.
- `preview_id` (`string`, required): Preview ID for save web source list.
- `mapping` (`object`, optional): Mapping for save web source list.

Example input: `{"import_id":"<import_id>","name":"<name>","preview_hash":"<preview_hash>","preview_id":"<preview_id>"}`

### `specstream_send_to_destination`

Send To Destination · Write

Send selected owned results with durable delivery identities.
Returns a JSON object with data.
Needs identity (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `destination_sha256` (`string`, required): Destination Sha256 for send to destination.
- `entity_ids` (`array`, required): Entity IDS for send to destination.
- `find_id` (`string`, required): Find ID for send to destination.
- `identity` (`string`, required): Identity for send to destination.

Example input: `{"destination_sha256":"<destination_sha256>","entity_ids":[],"find_id":"<find_id>","identity":"<identity>"}`

### `specstream_set_account_spend`

Set Account Spend · Write

Set an explicit owner authorization ceiling without funding or spending.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `spend_ceiling_cents` (`integer`, required): Spend Ceiling Cents for set account spend.

Example input: `{"spend_ceiling_cents":1}`

### `specstream_set_capacity_auto_topup`

Set Capacity Auto Topup · Write

Turn automatic top-up on or off and set its level, amount or search pack, and monthly limits (owner only).
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_set_capacity_budget`

Set Capacity Budget · Write

Set an owner-approved monthly account spending maximum.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `monthly_cents` (`integer`, required): Monthly Cents for set capacity budget.

Example input: `{"monthly_cents":1}`

### `specstream_share_find`

Share Find · Write

Snapshot share token.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for share find.
- `include_contacts` (`boolean`, optional): Include Contacts for share find. Default: `true`.

Example input: `{"id":"<id>"}`

### `specstream_start_capacity_auto_topup_card`

Start Capacity Auto Topup Card · Write

Open a Stripe page to save or change the card for automatic top-up, with no payment (owner only).
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs: none.

Example input: `{}`

### `specstream_start_capacity_request`

Start Capacity Request · Write

Ask for a list in plain words and run it within a maximum price, answering at once.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `confirm_max_cents` (`integer`, required): Confirm Max Cents for start capacity request.
- `idempotency_key` (`string`, required): Idempotency Key for start capacity request.
- `callback_on` (`string`, optional): "start" tells the callback when the run starts. "done" posts the finished rows to it when the run ends, as GET /v1/capacity/requests/{key}/rows answers them, in parts of at most 90 KB. Both tell it when nothing runs. One of: start, done. One of: `start`, `done`. Default: `start`.
- `callback_url` (`string`, optional): Callback URL for start capacity request.
- `fields` (`array`, optional): What to bring back for each row, in plain words, such as "headcount" or "Is it hiring for a finance role?". q can ask the same in one sentence.
- `locale` (`string`, optional): Locale for start capacity request. Default: `en-US`.
- `maximum_price_usd` (`number`, optional): Maximum Price Usd for start capacity request.
- `one_row_per` (`string`, optional): For the rows posted when done: one flat row per item of this list field, such as "people" for one person per row.
- `q` (`string`, optional): What to find, in plain words. With rows, what to bring back for each row.
- `requested_units` (`integer`, optional): Requested Units for start capacity request. Default: `25`.
- `rows` (`array`, optional): Companies to enrich, one object per row: the company name, website or domain, and any other columns that identify it. Each row is one result.

Example input: `{"confirm_max_cents":1,"idempotency_key":"<idempotency_key>"}`

### `specstream_start_crawl`

Start Crawl · Write

Start a crawl: read a site page by page, breadth first, as a job to poll. Each page that gives text is one page read.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `url` (`string`, required): URL for start crawl.
- `entire_domain` (`boolean`, optional): Entire Domain for start crawl. Default: `false`.
- `exclude_paths` (`array`, optional): Exclude Paths for start crawl.
- `format` (`string`, optional): Format for start crawl. One of: markdown, text, html. One of: `markdown`, `text`, `html`. Default: `markdown`.
- `idempotency_key` (`string`, optional): Your id for this call, at most 128 characters. The same id again answers with the same map charge or the same crawl job.
- `include` (`array`, optional): Include for start crawl.
- `include_paths` (`array`, optional): Include Paths for start crawl.
- `include_subdomains` (`boolean`, optional): Include Subdomains for start crawl. Default: `false`.
- `limit` (`integer`, optional): Limit for start crawl. Default: `100`.
- `max_depth` (`integer`, optional): Max Depth for start crawl.
- `sitemap` (`string`, optional): Sitemap for start crawl. One of: include, skip. One of: `include`, `skip`. Default: `include`.

Example input: `{"url":"<url>"}`

### `specstream_transfer_capacity_store_to_card`

Transfer Capacity Store To Card · Write

Authorize a recoverable Google Play to card billing transfer.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `idempotency_key` (`string`, required): Idempotency Key for transfer capacity store to card.
- `plan_id` (`string`, required): Plan ID for transfer capacity store to card.

Example input: `{"idempotency_key":"<idempotency_key>","plan_id":"<plan_id>"}`

### `specstream_update_account_key`

Update Account Key · Write

Turn zero retention on or off for an API key.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for update account key.
- `engine_rule` (`object`, optional): The engines a key's searches may ask (listed at /v1/search-engines): each item an
engine's name or the two-letter ISO 3166-1 code of the country its operator runs it from.
An engine is asked when ``allow`` is empty or names it or its country, and ``deny``
names neither.
- `eu_egress_only` (`boolean`, optional): Eu Egress Only for update account key.
- `expires_at` (`string`, optional): When the key stops working, a future time; null removes the end.
- `name` (`string`, optional): Name for update account key.
- `spend_cap_cents` (`integer`, optional): The most the key spends in total, in cents, counted over every call since the key was made. The account's balance and limits apply as well.
- `zero_retention` (`boolean`, optional): Zero Retention for update account key.

Example input: `{"id":"<id>"}`

### `specstream_update_cloud_import`

Update Cloud Import · Write

Start or stop daily source checks.
Returns a JSON object with data.
Needs import_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `import_id` (`string`, required): Import ID for update cloud import.
- `keepUpdated` (`boolean`, required): Keepupdated for update cloud import.

Example input: `{"import_id":"<import_id>","keepUpdated":true}`

### `specstream_update_destination_sync`

Update Destination Sync · Write

Start or stop continuous destination delivery.
Returns a JSON object with data.
Needs sync_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `keepSending` (`boolean`, required): Keepsending for update destination sync.
- `sync_id` (`string`, required): Sync ID for update destination sync.

Example input: `{"keepSending":true,"sync_id":"<sync_id>"}`

### `specstream_update_project`

Update Project · Write

Update a project only if the expected version matches.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for update project.
- `id` (`string`, required): ID for update project.
- `monthly_spend_cap_cents` (`integer`, optional): Monthly Spend Cap Cents for update project.
- `name` (`string`, optional): Name for update project.

Example input: `{"expected_version":1,"id":"<id>"}`

### `specstream_verify_capacity_store_purchase`

Verify Capacity Store Purchase · Write

Confirm an app store purchase before adding its balance.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `product_id` (`string`, required): Product ID for verify capacity store purchase.
- `provider` (`string`, required): Provider for verify capacity store purchase. One of: app_store, google_play. One of: `app_store`, `google_play`.
- `purchase_token` (`string`, required): Purchase Token for verify capacity store purchase.

Example input: `{"product_id":"<product_id>","provider":"app_store","purchase_token":"<purchase_token>"}`

### `specstream_workspace_add_note`

Workspace Add Note · Write

Add a versioned note to an authorized project or collection record.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace add note.
- `id` (`string`, required): ID for workspace add note.
- `text` (`string`, required): Text for workspace add note.
- `collection_id` (`string`, optional): Collection ID for workspace add note.
- `project_id` (`string`, optional): Project ID for workspace add note.
- `record_id` (`string`, optional): Record ID for workspace add note.

Example input: `{"expected_version":1,"id":"<id>","text":"<text>"}`

### `specstream_workspace_add_sso_domain`

Workspace Add SSO Domain · Write

Create an account-owned domain verification challenge.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `domain` (`string`, required): Domain for workspace add sso domain.

Example input: `{"domain":"<domain>"}`

### `specstream_workspace_combine_collections`

Workspace Combine Collections · Write

Combine your lists into one and keep each row's sources.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collection_ids` (`array`, required): Collection IDS for workspace combine collections.
- `name` (`string`, required): Name for workspace combine collections.
- `request_id` (`string`, required): Request ID for workspace combine collections.
- `project_id` (`string`, optional): Project ID for workspace combine collections.

Example input: `{"collection_ids":[],"name":"<name>","request_id":"<request_id>"}`

### `specstream_workspace_copy_collection`

Workspace Copy Collection · Write

Copy rows from a list you can open into a separate list you own.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace copy collection.
- `name` (`string`, required): Name for workspace copy collection.
- `request_id` (`string`, required): Request ID for workspace copy collection.
- `project_id` (`string`, optional): Project ID for workspace copy collection.

Example input: `{"collection_id":"<collection_id>","name":"<name>","request_id":"<request_id>"}`

### `specstream_workspace_edit_note`

Workspace Edit Note · Write

Edit a note you wrote with an expected version.
Returns a JSON object with data.
Needs note_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace edit note.
- `note_id` (`string`, required): Note ID for workspace edit note.
- `text` (`string`, required): Text for workspace edit note.
- `collection_id` (`string`, optional): Collection ID for workspace edit note.
- `project_id` (`string`, optional): Project ID for workspace edit note.
- `record_id` (`string`, optional): Record ID for workspace edit note.

Example input: `{"expected_version":1,"note_id":"<note_id>","text":"<text>"}`

### `specstream_workspace_invite_member`

Workspace Invite Member · Write

Invite a project member under current project management authority.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `email` (`string`, required): Email for workspace invite member.
- `project_id` (`string`, required): Project ID for workspace invite member.
- `request_id` (`string`, required): Request ID for workspace invite member.
- `expires_in_days` (`integer`, optional): Expires In Days for workspace invite member. One of: 1, 7, 30. One of: `1`, `7`, `30`. Default: `7`.
- `monthly_spend_cap_cents` (`integer`, optional): Monthly Spend Cap Cents for workspace invite member. Default: `0`.
- `role` (`string`, optional): Role for workspace invite member. One of: editor, viewer. One of: `editor`, `viewer`. Default: `editor`.

Example input: `{"email":"<email>","project_id":"<project_id>","request_id":"<request_id>"}`

### `specstream_workspace_name_passkey`

Workspace Name Passkey · Write

Name an account-owned passkey.
Returns a JSON object with data.
Needs credential_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `credential_id` (`string`, required): Credential ID for workspace name passkey.
- `label` (`string`, required): Label for workspace name passkey.

Example input: `{"credential_id":"<credential_id>","label":"<label>"}`

### `specstream_workspace_passkey_sign_in`

Workspace Passkey Sign In · Write

Verify a passkey assertion and establish a session.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `challenge_id` (`string`, required): Challenge ID for workspace passkey sign in.
- `credential` (`object`, required): Credential for workspace passkey sign in.

Example input: `{"challenge_id":"<challenge_id>","credential":{}}`

### `specstream_workspace_passkey_sign_in_options`

Workspace Passkey Sign In Options · Write

Start a browser-bound passkey sign-in ceremony.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs: none.

Example input: `{}`

### `specstream_workspace_read_notifications`

Workspace Read Notifications · Write

Mark the supplied authorized notifications read atomically.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `ids` (`array`, required): IDS for workspace read notifications.

Example input: `{"ids":[]}`

### `specstream_workspace_redeem_native_provider`

Workspace Redeem Native Provider · Write

Redeem verified provider proof for an app session or connection.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `code_verifier` (`string`, required): Code Verifier for workspace redeem native provider.
- `handoff_id` (`string`, required): Handoff ID for workspace redeem native provider.

Example input: `{"code_verifier":"<code_verifier>","handoff_id":"<handoff_id>"}`

### `specstream_workspace_register_passkey`

Workspace Register Passkey · Write

Verify and register a passkey for the current account.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `challenge_id` (`string`, required): Challenge ID for workspace register passkey.
- `credential` (`object`, required): Credential for workspace register passkey.
- `label` (`string`, required): Label for workspace register passkey.

Example input: `{"challenge_id":"<challenge_id>","credential":{},"label":"<label>"}`

### `specstream_workspace_register_passkey_options`

Workspace Register Passkey Options · Write

Start a browser-bound passkey registration ceremony.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs: none.

Example input: `{}`

### `specstream_workspace_register_push_device`

Workspace Register Push Device · Write

Register the current signed-in device for native notifications.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name for workspace register push device.
- `platform` (`string`, required): Platform for workspace register push device. One of: apns, fcm. One of: `apns`, `fcm`.
- `token` (`string`, required): Token for workspace register push device.

Example input: `{"name":"<name>","platform":"apns","token":"<token>"}`

### `specstream_workspace_register_web_push_subscription`

Workspace Register Web Push Subscription · Write

Register an encrypted browser push subscription for this signed-in account.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `endpoint` (`string`, required): Endpoint for workspace register web push subscription.
- `keys` (`object`, required): Keys for workspace register web push subscription.
- `name` (`string`, required): Name for workspace register web push subscription.
- `expirationTime` (`integer`, optional): Expirationtime for workspace register web push subscription.

Example input: `{"endpoint":"<endpoint>","keys":{},"name":"<name>"}`

### `specstream_workspace_report_result`

Workspace Report Result · Write

Report a billed result for review within its guarantee period.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `check` (`string`, required): Check for workspace report result.
- `collection_id` (`string`, required): Collection ID for workspace report result.
- `record_id` (`string`, required): Record ID for workspace report result.
- `request_id` (`string`, required): Request ID for workspace report result.
- `truth` (`string`, required): Truth for workspace report result.
- `link` (`string`, optional): Link for workspace report result.

Example input: `{"check":"<check>","collection_id":"<collection_id>","record_id":"<record_id>","request_id":"<request_id>","truth":"<truth>"}`

### `specstream_workspace_resolve_note`

Workspace Resolve Note · Write

Resolve or reopen an authorized note with an expected version.
Returns a JSON object with data.
Needs note_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace resolve note.
- `note_id` (`string`, required): Note ID for workspace resolve note.
- `resolved` (`boolean`, required): Resolved for workspace resolve note.
- `collection_id` (`string`, optional): Collection ID for workspace resolve note.
- `project_id` (`string`, optional): Project ID for workspace resolve note.
- `record_id` (`string`, optional): Record ID for workspace resolve note.

Example input: `{"expected_version":1,"note_id":"<note_id>","resolved":true}`

### `specstream_workspace_save_relationships`

Workspace Save Relationships · Write

Save a personal relationship column with an expected version.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace save relationships.
- `expected_version` (`integer`, required): Expected Version for workspace save relationships.
- `field_id` (`string`, required): Field ID for workspace save relationships.
- `label` (`string`, required): Label for workspace save relationships.

Example input: `{"collection_id":"<collection_id>","expected_version":1,"field_id":"<field_id>","label":"<label>"}`

### `specstream_workspace_save_view`

Workspace Save View · Write

Save a versioned personal view without changing source records.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace save view.
- `expected_version` (`integer`, required): Expected Version for workspace save view.
- `view` (`object`, required): View for workspace save view.

Example input: `{"collection_id":"<collection_id>","expected_version":1,"view":{}}`

### `specstream_workspace_set_schedule_delivery`

Workspace Set Schedule Delivery · Write

Set versioned monitor delivery settings without starting work.
Returns a JSON object with data.
Needs schedule_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace set schedule delivery.
- `schedule_id` (`string`, required): Schedule ID for workspace set schedule delivery.
- `notify` (`boolean`, optional): Notify for workspace set schedule delivery. Default: `true`.
- `sendTo` (`string`, optional): Sendto for workspace set schedule delivery.
- `slack` (`string`, optional): Slack for workspace set schedule delivery.
- `webhook` (`string`, optional): Webhook for workspace set schedule delivery.

Example input: `{"expected_version":1,"schedule_id":"<schedule_id>"}`

### `specstream_workspace_set_sso_policy`

Workspace Set SSO Policy · Write

Set SSO enforcement after a successful provider test.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace set sso policy.
- `require_sso` (`boolean`, required): Require SSO for workspace set sso policy.
- `scim_enabled` (`boolean`, required): SCIM Enabled for workspace set sso policy.

Example input: `{"domain_id":"<domain_id>","require_sso":true,"scim_enabled":true}`

### `specstream_workspace_set_your_field_value`

Workspace Set Your Field Value · Write

Set or clear one of your fields on selected records.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace set your field value.
- `expected_version` (`integer`, required): Expected Version for workspace set your field value.
- `key` (`string`, required): Key for workspace set your field value.
- `record_ids` (`array`, required): Record IDS for workspace set your field value.
- `value` (`string`, optional): Value for workspace set your field value.

Example input: `{"collection_id":"<collection_id>","expected_version":1,"key":"<key>","record_ids":[]}`

### `specstream_workspace_share_collection`

Workspace Share Collection · Write

Share a fixed copy of every record in a list through a link.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace share collection.
- `allow_export` (`boolean`, optional): Whether people with the link can download the rows. Default: `false`.
- `expires_days` (`integer`, optional): Days the link stays on. null keeps it on. Default: `90`.

Example input: `{"collection_id":"<collection_id>"}`

### `specstream_workspace_star_item`

Workspace Star Item · Write

Star or unstar currently authorized work with an expected version.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace star item.
- `item` (`object`, required): Item for workspace star item.
- `starred` (`boolean`, required): Starred for workspace star item.

Example input: `{"expected_version":1,"item":{},"starred":true}`

### `specstream_workspace_start_native_provider`

Workspace Start Native Provider · Write

Begin native provider authorization bound to an app-held PKCE secret.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `code_challenge` (`string`, required): Code Challenge for workspace start native provider.
- `provider` (`string`, required): Provider for workspace start native provider.
- `email` (`string`, optional): Email for workspace start native provider.
- `enroll` (`boolean`, optional): Enroll for workspace start native provider. Default: `false`.
- `purpose` (`string`, optional): Purpose for workspace start native provider. One of: signin, connection. One of: `signin`, `connection`. Default: `signin`.
- `ref` (`string`, optional): Ref for workspace start native provider.

Example input: `{"code_challenge":"<code_challenge>","provider":"<provider>"}`

### `specstream_workspace_start_placement`

Workspace Start Placement · Write

Approve the placement quote and enqueue durable background placement.
Returns a JSON object with data.
Needs import_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `confirm_max_cents` (`integer`, required): Confirm Max Cents for workspace start placement.
- `import_id` (`string`, required): Import ID for workspace start placement.
- `quote_hash` (`string`, required): Quote Hash for workspace start placement.
- `quote_id` (`string`, required): Quote ID for workspace start placement.
- `request_id` (`string`, required): Request ID for workspace start placement.

Example input: `{"confirm_max_cents":1,"import_id":"<import_id>","quote_hash":"<quote_hash>","quote_id":"<quote_id>","request_id":"<request_id>"}`

### `specstream_workspace_start_sso`

Workspace Start SSO · Write

Start a single-use browser-bound SAML sign-in.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `domain` (`string`, required): Domain for workspace start sso.
- `enroll` (`boolean`, optional): Enroll for workspace start sso. Default: `false`.

Example input: `{"domain":"<domain>"}`

### `specstream_workspace_update_notification`

Workspace Update Notification · Write

Persist read or archive state from your signed-in browser.
Returns a JSON object with data.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `identifier` (`string`, required): Identifier for workspace update notification.
- `archived` (`boolean`, optional): Archived for workspace update notification.
- `read` (`boolean`, optional): Read for workspace update notification.

Example input: `{"identifier":"<identifier>"}`

### `specstream_workspace_verify_sso_domain`

Workspace Verify SSO Domain · Write

Verify the account's domain through its DNS TXT record.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace verify sso domain.

Example input: `{"domain_id":"<domain_id>"}`

### `specstream_cancel_capacity_funding`

Cancel Capacity Funding · Destructive

Close an unpaid funding request without duplicating payment.
Returns a JSON object with data.
Needs funding_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `funding_id` (`string`, required): Funding ID for cancel capacity funding.

Example input: `{"funding_id":"<funding_id>"}`

### `specstream_cancel_crawl`

Cancel Crawl · Destructive

Stop a crawl job; the pages it read stay.
Returns a JSON object with data.
Needs crawl_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `crawl_id` (`string`, required): Crawl ID for cancel crawl.

Example input: `{"crawl_id":"<crawl_id>"}`

### `specstream_cancel_find`

Cancel Find · Destructive

Request cancel.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for cancel find.

Example input: `{"id":"<id>"}`

### `specstream_cancel_investigation_batch`

Cancel Investigation Batch · Destructive

Stop searches that haven't started and cancel running ones where possible. Finished results stay yours.
Returns a JSON object with data.
Needs batch_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `batch_id` (`string`, required): Batch ID for cancel investigation batch.

Example input: `{"batch_id":"<batch_id>"}`

### `specstream_cancel_research`

Cancel Research · Destructive

Stop a research run at its next step; a cancelled run is not charged.
Returns a JSON object with data.
Needs run_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `run_id` (`string`, required): Run ID for cancel research.

Example input: `{"run_id":"<run_id>"}`

### `specstream_delete_account`

Delete Account · Destructive

Revoke and grace purge.
Returns a JSON object with data.

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs: none.

Example input: `{}`

### `specstream_disconnect_connection`

Disconnect Connection · Destructive

Revoke a connected provider.
Returns a JSON object with data.
Needs identity (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `identity` (`string`, required): Identity for disconnect connection.

Example input: `{"identity":"<identity>"}`

### `specstream_patch_research_rule`

Patch Research Rule · Destructive

Pause, resume or delete a rule. Pausing holds queued searches; running ones finish. Resuming never raises the limits.
Returns a JSON object with data.
Needs rule_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for patch research rule.
- `rule_id` (`string`, required): Rule ID for patch research rule.
- `state` (`string`, required): State for patch research rule. One of: active, paused, deleted. One of: `active`, `paused`, `deleted`.

Example input: `{"expected_version":1,"rule_id":"<rule_id>","state":"active"}`

### `specstream_remove_capacity_auto_topup_card`

Remove Capacity Auto Topup Card · Destructive

Remove the saved card; automatic top-up turns off (owner only).
Returns a JSON object with data.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_revoke_account_key`

Revoke Account Key · Destructive

Revoke an API key.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for revoke account key.

Example input: `{"id":"<id>"}`

### `specstream_revoke_capacity_authority`

Revoke Capacity Authority · Destructive

Revoke a delegated spending limit without removing saved results.
Returns a JSON object with data.
Needs authority_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `authority_id` (`string`, required): Authority ID for revoke capacity authority.

Example input: `{"authority_id":"<authority_id>"}`

### `specstream_revoke_customer_webhook`

Revoke Customer Webhook · Destructive

Revoke an owned callback and cancel queued deliveries.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): ID for revoke customer webhook.

Example input: `{"id":"<id>"}`

### `specstream_revoke_project_member`

Revoke Project Member · Destructive

Revoke project membership without deleting owned work.
Returns a JSON object with data.
Needs id, account_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `account_id` (`string`, required): Account ID for revoke project member.
- `id` (`string`, required): ID for revoke project member.

Example input: `{"account_id":"<account_id>","id":"<id>"}`

### `specstream_revoke_share`

Revoke Share · Destructive

Revoke snapshot.
Returns a JSON object with data.
Needs id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for revoke share.
- `share_id` (`string`, required): Share ID for revoke share.

Example input: `{"id":"<id>","share_id":"<share_id>"}`

### `specstream_revoke_spec_link`

Revoke Spec Link · Destructive

Turn off a saved-request link without changing the request or its results.
Returns a JSON object with data.
Needs id, link_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `id` (`string`, required): ID for revoke spec link.
- `link_id` (`string`, required): Link ID for revoke spec link.

Example input: `{"id":"<id>","link_id":"<link_id>"}`

### `specstream_update_capacity_schedule`

Update Capacity Schedule · Destructive

Pause, resume, or delete recurring work without buying missed occurrences.
Returns a JSON object with data.
Needs schedule_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `expected_version` (`integer`, required): Expected Version for update capacity schedule.
- `schedule_id` (`string`, required): Schedule ID for update capacity schedule.
- `state` (`string`, required): State for update capacity schedule. One of: active, paused, deleted. One of: `active`, `paused`, `deleted`.

Example input: `{"expected_version":1,"schedule_id":"<schedule_id>","state":"active"}`

### `specstream_workspace_configure_sso`

Workspace Configure SSO · Destructive

Configure verified SAML metadata for an owned domain.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace configure sso.
- `metadata_url` (`string`, required): Metadata URL for workspace configure sso.
- `provider` (`string`, required): Provider for workspace configure sso. One of: Okta, Microsoft Entra ID, Google Workspace, OneLogin, JumpCloud, Another SAML 2.0 provider. One of: `Okta`, `Microsoft Entra ID`, `Google Workspace`, `OneLogin`, `JumpCloud`, `Another SAML 2.0 provider`.

Example input: `{"domain_id":"<domain_id>","metadata_url":"<metadata_url>","provider":"Okta"}`

### `specstream_workspace_delete_note`

Workspace Delete Note · Destructive

Remove an authorized note with an expected version.
Returns a JSON object with data.
Needs note_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `expected_version` (`integer`, required): Expected Version for workspace delete note.
- `note_id` (`string`, required): Note ID for workspace delete note.
- `collection_id` (`string`, optional): Collection ID for workspace delete note.
- `project_id` (`string`, optional): Project ID for workspace delete note.
- `record_id` (`string`, optional): Record ID for workspace delete note.

Example input: `{"expected_version":1,"note_id":"<note_id>"}`

### `specstream_workspace_delete_passkey`

Workspace Delete Passkey · Destructive

Revoke an account-owned passkey.
Returns a JSON object with data.
Needs credential_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `credential_id` (`string`, required): Credential ID for workspace delete passkey.

Example input: `{"credential_id":"<credential_id>"}`

### `specstream_workspace_delete_photo`

Workspace Delete Photo · Destructive

Remove the current account's private profile photo.
Returns a JSON object with data.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `specstream_workspace_revoke_push_device`

Workspace Revoke Push Device · Destructive

Stop notifications to a registered device.
Returns a JSON object with data.
Needs device_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `device_id` (`string`, required): Device ID for workspace revoke push device.

Example input: `{"device_id":"<device_id>"}`

### `specstream_workspace_revoke_scim_token`

Workspace Revoke SCIM Token · Destructive

Revoke a domain's provisioning credential.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace revoke scim token.

Example input: `{"domain_id":"<domain_id>"}`

### `specstream_workspace_rotate_scim_token`

Workspace Rotate SCIM Token · Destructive

Rotate a verified domain's provisioning credential.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace rotate scim token.

Example input: `{"domain_id":"<domain_id>"}`

### `specstream_workspace_save_your_fields`

Workspace Save Your Fields · Destructive

Add, rename or remove your own fields on a collection.
Returns a JSON object with data.
Needs collection_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `collection_id` (`string`, required): Collection ID for workspace save your fields.
- `expected_version` (`integer`, required): Expected Version for workspace save your fields.
- `fields` (`array`, required): Fields for workspace save your fields.

Example input: `{"collection_id":"<collection_id>","expected_version":1,"fields":[]}`

### `specstream_workspace_set_scim_mappings`

Workspace Set SCIM Mappings · Destructive

Set domain-to-project mappings under current owner authority.
Returns a JSON object with data.
Needs domain_id (identifiers returned by earlier SpecStream calls).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `domain_id` (`string`, required): Domain ID for workspace set scim mappings.
- `mappings` (`array`, required): Mappings for workspace set scim mappings.

Example input: `{"domain_id":"<domain_id>","mappings":[]}`

## Related

Other Search connectors ([all 89](/agentkit/connectors/?category=search)):

- [Supadata](/agentkit/connectors/supadata/): Scalekit connector, API key, 21 tools
- [Exa](/agentkit/connectors/exa/): Scalekit connector, API key, 21 tools
- [Brave Search](/agentkit/connectors/brave/): Scalekit connector, API key, 18 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
