> **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 the Typefully MCP server

The Typefully MCP connector routes your AI agent's tool calls to Typefully's own MCP server through Scalekit. Each user signs in to Typefully once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 24 tools.

**Authentication:** OAuth 2.1/DCR
**Categories:** Marketing
**Tools:** 24: 11 read, 4 write, 9 destructive
**Users sign in with:** OAuth
**OAuth app:** Your own Typefully MCP server app
**Built by:** Vendor MCP
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=TYPEFULLYMCP)

## 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 Typefully MCP connection

   In **AgentKit > Connections**, create a Typefully MCP connection and copy its redirect URI. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Register an OAuth app

   Typefully MCP server connections use your own OAuth app. Register one with Typefully MCP server and add the redirect URI you copied.

   Then enter the app's Client ID and Client Secret on the Typefully MCP connection.

5. ### 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 = 'typefullymcp'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Typefully MCP:', 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: 'typefullymcp_get_me',
     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 = "typefullymcp"
   identifier = "user_123"

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

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

   ```bash
   python quickstart.py
   ```

   Each user signs in once. 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: 'typefullymcp_get_draft',
  toolInput: {
    draft_id: 1,
    social_set_id: 1,
  },
  connector: 'typefullymcp',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="typefullymcp_get_draft",
    tool_input={
        "draft_id": 1,
        "social_set_id": 1,
    },
    connection_name="typefullymcp",
    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.

### `typefullymcp_get_draft`

Get draft · Read-only

Retrieve a specific draft by ID, including its content for all configured platforms, status, and scheduling information.

If the draft has comment threads, the response includes Typefully comment-thread markers in `posts[*].text` and, for X Articles, in `platforms.x_article.content_markdown`. These markers are structural anchor metadata for `GET → modify → PATCH` round-trips; preserve them exactly when editing.

For read-only display/export, pass `?exclude_comment_markers=true` to render draft text without markers. Content returned with that flag set should not be PATCHed back unless you intend to resolve or remove comment anchors.

**Draft statuses:** 'draft' = saved but not scheduled. 'scheduled' = queued to auto-publish at its scheduled_date. 'planned' = dated but inert: it has a scheduled_date but will NOT auto-publish until confirmed (by setting publish_at). A planned draft whose scheduled_date has passed is NOT overdue and NOT a failure - it simply hasn't been confirmed; replan it or confirm it. 'publishing' = a publish is in flight (transient). 'published' = successfully posted. 'error' = publishing failed.

**Required permission:** READ access to this social set.

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

Inputs:

- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.
- `exclude_comment_markers` (`boolean`, optional): When true, render `posts[*].text` as plain user-visible text without `<typ:comment-thread>` markers, and render X Article `content_markdown` without comment markers. Use only for read-only flows (LLM context windows, exports). The default (false) emits markers so a round-trip back to PATCH preserves comment anchors. Default: `false`.

Example input: `{"draft_id":1,"social_set_id":1}`

### `typefullymcp_get_me`

Get current user · Read-only

Retrieve the currently authenticated Typefully user associated with your API Key

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

Inputs: none.

Example input: `{}`

### `typefullymcp_get_media_status`

Get media status · Read-only

Retrieves the processing status of an uploaded media file. Poll this endpoint after uploading to check when the file is ready to use in drafts.

If no file is received before the upload URL expires (1 hour), the media transitions to 'failed' — create a new media upload and try again.

**Required permission:** READ access to the social set.

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

Inputs:

- `media_id` (`string`, required): No description.
- `social_set_id` (`number`, required): No description.

Example input: `{"media_id":"<media_id>","social_set_id":1}`

### `typefullymcp_get_queue`

Get queue · Read-only

Retrieve queue slots and the scheduled AND planned drafts occupying them, between `start_date` and `end_date` (inclusive). Check each draft's `status`: 'scheduled' = will auto-publish at its `scheduled_date`; 'planned' = dated but inert - it will NOT auto-publish until confirmed (a past planned date is not overdue and not a failure).

**Required permission:** READ access to this social set.

Notes:
- `start_date` and `end_date` are interpreted in the social set timezone.
- Ranges larger than 62 days are rejected.

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

Inputs:

- `end_date` (`string`, required): No description.
- `social_set_id` (`number`, required): No description.
- `start_date` (`string`, required): No description.

Example input: `{"end_date":"<end_date>","social_set_id":1,"start_date":"<start_date>"}`

### `typefullymcp_get_queue_schedule`

Get queue schedule · Read-only

Retrieve the queue schedule rules for a social set.

**Required permission:** READ access to this social set.

Behavior:
- If the schedule row does not exist yet, it is created with defaults.

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

Inputs:

- `social_set_id` (`number`, required): No description.

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

### `typefullymcp_get_social_set_details`

Get social set details · Read-only

Retrieve detailed information about a social set, including every configured social media platform (X, LinkedIn, Mastodon, Threads, Bluesky) with account details and profile information.

**Required permission:** READ access to the social set.

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

Inputs:

- `social_set_id` (`number`, required): No description.

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

### `typefullymcp_linkedin_resolve_linkedin_organization_from_url`

Resolve LinkedIn organization · Read-only

Resolve a LinkedIn company/school URL into organization metadata that can be used to build LinkedIn mention syntax in post text.

This endpoint is resolver-only and is not a general organization search endpoint.

Mention format: `@[Company Name](urn:li:organization:123456)`

**Required permission:** READ access to the social set.

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

Inputs:

- `organization_url` (`string`, required): Public LinkedIn organization URL (company or school profile).
- `social_set_id` (`number`, required): No description.

Example input: `{"organization_url":"<organization_url>","social_set_id":1}`

### `typefullymcp_list_comments`

List comment threads · Read-only

Retrieve comment threads attached to a draft, ordered by creation time. Each thread includes the original `selected_text` snapshot and the full ordered list of comments.

By default this endpoint returns only unresolved threads (`status=unresolved`). Use `status=resolved` to list resolved threads or `status=all` to list both unresolved and resolved threads. Example: `GET /v2/social-sets/4/drafts/12/comment-threads?status=all&limit=50`.

**Required permission:** READ access to this social set.

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

Inputs:

- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.
- `limit` (`number`, optional): No description. Default: `10`.
- `offset` (`number`, optional): No description. Default: `0`.
- `platform` (`string`, optional): Optional platform filter.
- `status` (`string`, optional): Resolution filter. Defaults to `unresolved`, so resolved threads are omitted unless you request `resolved` or `all`. One of: `unresolved`, `resolved`, `all`. Default: `unresolved`.

Example input: `{"draft_id":1,"social_set_id":1}`

### `typefullymcp_list_drafts`

List drafts · Read-only

Retrieve all drafts for a specific social set with optional filtering and sorting. Drafts are ordered by last edited date (most recent first) by default.

**Draft statuses:** 'draft' = saved but not scheduled. 'scheduled' = queued to auto-publish at its scheduled_date. 'planned' = dated but inert: it has a scheduled_date but will NOT auto-publish until confirmed (by setting publish_at). A planned draft whose scheduled_date has passed is NOT overdue and NOT a failure - it simply hasn't been confirmed; replan it or confirm it. 'publishing' = a publish is in flight (transient). 'published' = successfully posted. 'error' = publishing failed.

**Required permission:** READ access to this social set.

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

Inputs:

- `social_set_id` (`number`, required): No description.
- `limit` (`string`, optional): Maximum number of items to return per page. Default: `10`.
- `offset` (`number`, optional): Number of items to skip from the beginning. Default: `0`.
- `order_by` (`string`, optional): Allowed order_by fields for draft listing - prevents SQL injection. One of: `created_at`, `-created_at`, `updated_at`, `-updated_at`, `scheduled_date`, `-scheduled_date`, `published_at`, `-published_at`. Default: `-updated_at`.
- `status` (`string`, optional): Filter by draft status. 'draft' = saved but not scheduled. 'scheduled' = queued to auto-publish at its scheduled_date. 'planned' = dated but inert: it has a scheduled_date but will NOT auto-publish until confirmed (by setting publish_at). A planned draft whose scheduled_date has passed is NOT overdue and NOT a failure - it simply hasn't been confirmed; replan it or confirm it. 'publishing' = a publish is in flight (transient). 'published' = successfully posted. 'error' = publishing failed.
- `tag` (`string`, optional): No description.

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

### `typefullymcp_list_social_sets`

List social sets · Read-only

Retrieve all social sets (accounts) you can access. This includes accounts you own directly and accounts that belong to teams you are a member of.

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

Inputs:

- `limit` (`string`, optional): Maximum number of items to return per page. Default: `10`.
- `offset` (`number`, optional): Number of items to skip from the beginning. Default: `0`.

Example input: `{}`

### `typefullymcp_list_tags`

List tags · Read-only

Retrieve all tags for a social set, ordered by their slugs.

**Required permission:** READ access to the social set.

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

Inputs:

- `social_set_id` (`number`, required): No description.
- `limit` (`string`, optional): Maximum number of items to return per page. Default: `10`.
- `offset` (`number`, optional): Number of items to skip from the beginning. Default: `0`.

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

### `typefullymcp_comments_add_comment_to_thread`

Add comment · Write

Append a comment to an existing comment thread. Returns the full updated thread.

**Required permission:** WRITE access to this social set.

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

Inputs:

- `comment_thread_id` (`string`, required): No description.
- `draft_id` (`number`, required): No description.
- `requestBody` (`object`, required): Add a comment to an existing comment thread.
- `social_set_id` (`number`, required): No description.

Example input: `{"comment_thread_id":"<comment_thread_id>","draft_id":1,"requestBody":{},"social_set_id":1}`

### `typefullymcp_create_comment`

Create comment thread · Write

Create a new comment thread on a draft. A comment thread is the anchored discussion container; comments are the messages inside the thread.

For post platforms (`x`, `linkedin`, `mastodon`, `threads`, `bluesky`, `substack`), provide `post_index` and anchor to a substring of that post's visible text using `selected_text`. Use `occurrence` to disambiguate when the same substring appears more than once; omit it to anchor on the first match.

For X Articles, send `platform: "x_article"` and omit `post_index` (or use 0). `selected_text` matches commentable article text in document order; Markdown syntax, media tags, X post embed tags, and fenced-code text are ignored. Code blocks do not support comment anchors. X Article text comments cannot overlap existing stored X Article anchors; unsupported or overlapping selections return `400 VALIDATION_ERROR`.

LinkedIn mentions appear in `posts[*].text` as `@[Name](urn:li:organization:ID)` or `@[Name](urn:li:person:ID)`. A mention is indivisible — `selected_text` must either contain the entire mention substring or stay outside it.

**Required permission:** WRITE access to this social set.

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

Inputs:

- `draft_id` (`number`, required): No description.
- `requestBody` (`string`, required): The JSON request body.
- `social_set_id` (`number`, required): No description.

Example input: `{"draft_id":1,"requestBody":"<requestBody>","social_set_id":1}`

### `typefullymcp_create_media_upload`

Create media upload · Write

Generate a presigned S3 upload URL for images, videos, GIFs, or PDFs. After you receive the URL, upload the file contents with a PUT request and then reference the returned media_id when creating drafts.

**Uploading:** Send a plain PUT with only raw file bytes as the body — no extra headers (`Content-Type`, `Authorization`, etc.). The presigned URL signature was calculated without them, so adding headers causes a `403 SignatureDoesNotMatch`. Use `curl -T <file>` (not `--data-binary`), `requests.put(url, data=file_bytes)` in Python, or `fetch(url, {method:'PUT', body:buffer})` in JS. A successful upload returns `200` or `204`.

**Required permission:** WRITE access to the social set.

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

Inputs:

- `requestBody` (`object`, required): Schema for media upload request
- `social_set_id` (`number`, required): No description.

Example input: `{"requestBody":{},"social_set_id":1}`

### `typefullymcp_create_tag`

Create tag · Write

Create a new tag for a social set. The slug is automatically generated from the tag name, which must be unique per social set.

**Required permission:** WRITE access to the social set.

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

Inputs:

- `requestBody` (`object`, required): Request schema for creating a tag
- `social_set_id` (`number`, required): No description.

Example input: `{"requestBody":{},"social_set_id":1}`

### `typefullymcp_comments_resolve_thread`

Resolve comment thread · Destructive

Resolves the comment thread and removes the corresponding comment markers from the text.

**Required permission:** Either authorship of the comment thread or WRITE access on the social set.

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

Inputs:

- `comment_thread_id` (`string`, required): No description.
- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.

Example input: `{"comment_thread_id":"<comment_thread_id>","draft_id":1,"social_set_id":1}`

### `typefullymcp_create_draft`

Create draft · Destructive

Create a new draft with content for one or more social media platforms. The draft can be saved as a draft, planned (`plan_at`: dated but inert - it never auto-publishes until confirmed), scheduled for later publishing, or published immediately.

**Account-level settings:** This endpoint automatically applies the following account-level settings if enabled: Auto-Retweet, Auto-Plug, and Natural Posting Time.

**Required permission:** WRITE access to create drafts (planning included - it creates no publishing commitment). PUBLISH access is required to schedule or publish immediately.

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

Inputs:

- `requestBody` (`object`, required): Request schema for creating a draft
- `social_set_id` (`number`, required): No description.

Example input: `{"requestBody":{},"social_set_id":1}`

### `typefullymcp_delete_comment`

Delete comment · Destructive

Deletes a comment from a comment thread. If `comment_id` identifies the *root* (oldest) comment, the entire thread is deleted and the corresponding comment markers are removed from the text.

**Required permission:** Authorship of the comment or WRITE access on the social set.

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

Inputs:

- `comment_id` (`string`, required): No description.
- `comment_thread_id` (`string`, required): No description.
- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.

Example input: `{"comment_id":"<comment_id>","comment_thread_id":"<comment_thread_id>","draft_id":1,"social_set_id":1}`

### `typefullymcp_delete_draft`

Delete draft · Destructive

Delete a draft. Requires WRITE access. You can delete your own drafts in any status (DRAFT, ERROR, SCHEDULED, PUBLISHED, PUBLISHING) with WRITE access. Drafts created by other users also require WRITE access to delete.

**Required permission:** WRITE access to this social set.

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

Inputs:

- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.

Example input: `{"draft_id":1,"social_set_id":1}`

### `typefullymcp_delete_thread`

Delete comment thread · Destructive

Deletes the comment thread along with all its comments and removes the corresponding comment markers from the text.

**Required permission:** Authorship of the comment thread or WRITE access on the social set.

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

Inputs:

- `comment_thread_id` (`string`, required): No description.
- `draft_id` (`number`, required): No description.
- `social_set_id` (`number`, required): No description.

Example input: `{"comment_thread_id":"<comment_thread_id>","draft_id":1,"social_set_id":1}`

### `typefullymcp_edit_draft`

Update draft · Destructive

Update an existing draft with partial update semantics. Only provided fields are updated; omitted fields remain unchanged. Scheduled drafts require publish access to edit.

## Planning

A planned draft is dated but inert: it has a `scheduled_date` but never auto-publishes until confirmed. Use `plan_at` and `publish_at` (mutually exclusive per request) to move a draft through the lifecycle:

- `plan_at=<datetime|"next-free-slot">` on a plain draft plans it, on a planned draft moves the date (both write access), and on a scheduled draft unschedules it into a plan (publish access - it disarms a live schedule).
- `publish_at=<datetime|"next-free-slot">` on a planned draft confirms it into a real schedule (publish access; echo the draft's `scheduled_date` to confirm at the planned date). `publish_at="now"` publishes it immediately.
- An explicit `plan_at=null` or `publish_at=null` clears the date and returns the draft to plain draft status (write access from planned, publish access from scheduled).
- Content edits on a planned draft require write access only.

## Note about Comment-thread markers

If the draft has comment threads, submitted `posts[*].text` and X Article `platforms.x_article.content_markdown` must preserve the Typefully comment-thread markers received from `GET /drafts/{id}`. Validation is platform-level: every comment thread anchored on a platform must appear somewhere in that platform's submitted text.

Recommended edit flow: GET the draft without `exclude_comment_markers`, modify text while preserving markers exactly, then PATCH with `force_overwrite_comments: false` (the default).

- `409 COMMENTS_MARKER_MISMATCH` will be thrown if an expected comment thread marker is missing unless `"force_overwrite_comments": true` is set, in which case the affected threads are resolved server-side.
- `400 COMMENTS_MARKER_UNKNOWN_ID` will be thrown if you submit an id that doesn't exist on this draft. - `400 COMMENTS_MARKER_MALFORMED` will be thrown if the marker tag is malformed (bad UUID, unbalanced, attribute violations, etc.).

Pass `?exclude_comment_markers=true` to render the response text without markers (read-only / display rendering — does NOT skip server-side marker validation on the request body). Do not PATCH content returned with that flag unless you intend to resolve or remove comment anchors.

**Required permission:** WRITE access to edit drafts. PUBLISH access is required to edit scheduled drafts, schedule, or publish.

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

Inputs:

- `draft_id` (`number`, required): No description.
- `requestBody` (`object`, required): Request schema for updating a draft (partial updates supported)
- `social_set_id` (`number`, required): No description.
- `exclude_comment_markers` (`boolean`, optional): Render the response's `posts[*].text` as plain text without `<typ:comment-thread>` markers, and render X Article `content_markdown` without comment markers. Render-only — does not affect request-body validation. Default: `false`.

Example input: `{"draft_id":1,"requestBody":{},"social_set_id":1}`

### `typefullymcp_queue_put_queue_schedule`

Replace queue schedule · Destructive

Replace the queue schedule rules for a social set.

**Required permission:** ADMIN access to this social set.
Semantics: full replacement (atomic).

Rule validation:
- `h` in `0..23`, `m` in `0..59`
- `days` values are one of: `mon,tue,wed,thu,fri,sat,sun`
- Duplicate day+time combinations are rejected

Note: `rules=[]` is allowed and represents an empty schedule.

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

Inputs:

- `requestBody` (`object`, required): Request schema for fully replacing queue schedule rules.
- `social_set_id` (`number`, required): No description.

Example input: `{"requestBody":{},"social_set_id":1}`

### `typefullymcp_update_comment`

Update comment · Destructive

Update the plain-text body of a single comment. Returns the full updated thread.

**Required permission:** Authorship of the comment.

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

Inputs:

- `comment_id` (`string`, required): No description.
- `comment_thread_id` (`string`, required): No description.
- `draft_id` (`number`, required): No description.
- `requestBody` (`object`, required): Update the text body of a single comment within a comment thread.
- `social_set_id` (`number`, required): No description.

Example input: `{"comment_id":"<comment_id>","comment_thread_id":"<comment_thread_id>","draft_id":1,"requestBody":{},"social_set_id":1}`

### `typefullymcp_update_media`

Update media · Destructive

Updates metadata of an uploaded media file. Currently supports setting or clearing `alt_text`.

Alt text is copied into a draft when the media is attached (draft create or edit), so set it **before** referencing the media in a draft. To reliably change alt text on media already attached to a draft, update the media and then re-send the draft content with a draft PATCH. (Some platforms fall back to the media's current alt text when the draft snapshot has none, but only re-sending the draft content updates it everywhere.)

**Required permission:** WRITE access to the social set.

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

Inputs:

- `media_id` (`string`, required): No description.
- `requestBody` (`object`, required): Schema for media update request
- `social_set_id` (`number`, required): No description.

Example input: `{"media_id":"<media_id>","requestBody":{},"social_set_id":1}`

## Related

Other Marketing connectors ([all 190](/agentkit/connectors/?category=marketing)):

- [HubSpot](/agentkit/connectors/hubspot/): Scalekit connector, OAuth, 457 tools
- [Mailchimp](/agentkit/connectors/mailchimp/): Scalekit connector, OAuth, 73 tools
- [Twitter / X](/agentkit/connectors/twitteroauth/): Scalekit connector, OAuth, 92 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 |
