> **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 Gladia

The Gladia connector lets your AI agent act in each user's Gladia account. Each user connects their own Gladia API key once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 8 tools.

**Authentication:** API Key
**Categories:** Transcription, AI
**Tools:** 8: 4 read, 2 write, 2 destructive
**Users sign in with:** API key
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=GLADIA)

## 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 Gladia connection

   In **AgentKit > Connections**, create a Gladia 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 = 'gladia'
   const identifier = 'user_123'

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

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

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

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their Gladia 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: 'gladia_live_session_get',
  toolInput: {
    job_id: '<job_id>',
  },
  connector: 'gladia',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="gladia_live_session_get",
    tool_input={
        "job_id": "<job_id>",
    },
    connection_name="gladia",
    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.

### `gladia_live_session_get`

Get Live Transcription · Read-only

Get a live transcription session with its status, parameters and final result.
Returns status, the session parameters and result with the transcript and any post-processing outputs.
Use this after a live session ends to read what was said.

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

Inputs:

- `job_id` (`string`, required): ID of the live transcription job, returned when it was created or by the list tool. Example: 45463597-20b7-4af7-b3b3-f5fb778203ab.

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

### `gladia_live_sessions_list`

List Live Transcriptions · Read-only

List live (real-time) transcription sessions, filtered by date or status.
Returns items with each session's id, status, created_at and parameters, plus pagination links.
Use this to find a past live session. Use gladia_live_session_get for its transcript.

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

Inputs:

- `after_date` (`string`, optional): Only jobs after this date, ISO 8601. Combine with before_date for a range. Example: 2026-11-01T00:00:00Z.
- `before_date` (`string`, optional): Only jobs before this date, ISO 8601. Example: 2026-11-30T23:59:59Z.
- `date` (`string`, optional): Only jobs from this day, YYYY-MM-DD. Example: 2026-11-01.
- `limit` (`integer`, optional): Maximum jobs to return. Defaults to 20. Example: 20.
- `offset` (`integer`, optional): Number of jobs to skip, for paging. Defaults to 0. Example: 0.
- `status` (`string`, optional): Only jobs with this status. Example: done. One of: `queued`, `processing`, `done`, `error`.

Example input: `{}`

### `gladia_transcription_get`

Get Transcription · Read-only

Get a pre-recorded transcription job with its status and, when done, its full result.
Returns status (queued, processing, done or error), file metadata, and result with the transcript (full text and utterances) plus any summary, translation, subtitles and other add-on outputs.
Use this to poll a job from gladia_transcription_create.

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

Inputs:

- `job_id` (`string`, required): ID of the pre-recorded transcription job, returned when it was created or by the list tool. Example: 45463597-20b7-4af7-b3b3-f5fb778203ab.

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

### `gladia_transcriptions_list`

List Transcriptions · Read-only

List pre-recorded transcription jobs, newest first, filtered by date or status.
Returns items with each job's id, status, created_at, file metadata and request parameters, plus pagination links.
Use this to find a job id. Use gladia_transcription_get for the transcript.

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

Inputs:

- `after_date` (`string`, optional): Only jobs after this date, ISO 8601. Combine with before_date for a range. Example: 2026-11-01T00:00:00Z.
- `before_date` (`string`, optional): Only jobs before this date, ISO 8601. Example: 2026-11-30T23:59:59Z.
- `date` (`string`, optional): Only jobs from this day, YYYY-MM-DD. Example: 2026-11-01.
- `limit` (`integer`, optional): Maximum jobs to return. Defaults to 20. Example: 20.
- `offset` (`integer`, optional): Number of jobs to skip, for paging. Defaults to 0. Example: 0.
- `status` (`string`, optional): Only jobs with this status. Example: done. One of: `queued`, `processing`, `done`, `error`.

Example input: `{}`

### `gladia_audio_url_upload`

Import Audio from URL · Write

Import an audio or video file from a public URL into Gladia storage for later transcription.
Returns audio_url (a Gladia file URL) and audio_metadata (id, filename, duration, channels).
Optional: gladia_transcription_create also accepts external URLs directly. Use this to check a file's metadata or reuse it across jobs.

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

Inputs:

- `audio_url` (`string`, required): Public URL of the audio or video file to import. Example: https://files.gladia.io/example/audio-transcription/split_infinity.wav.

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

### `gladia_transcription_create`

Start Transcription · Write

Start an asynchronous transcription of a pre-recorded audio or video file from a URL, with optional diarization, translation, summary, subtitles, sentiment, PII redaction and LLM prompts.
Returns the job id and a result_url.
Poll gladia_transcription_get until status is done to read the transcript.

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

Inputs:

- `audio_url` (`string`, required): URL of the file: a Gladia file URL from gladia_audio_url_upload or any public audio or video URL. Example: https://files.gladia.io/example/audio-transcription/split_infinity.wav.
- `audio_to_llm` (`boolean`, optional): Run LLM prompts over the transcript. Example: false.
- `audio_to_llm_config` (`object`, optional): LLM settings when audio_to_llm is on: prompts (list of instructions) and optional model. Example: {"prompts": ["Extract the key points from the transcription"]}
- `callback` (`boolean`, optional): Send the result to callback_config.url when the job ends. Example: false.
- `callback_config` (`object`, optional): Callback settings when callback is on: url and method (POST or PUT). Example: {"url": "https://your-server.com/gladia-callback", "method": "POST"}
- `custom_metadata` (`object`, optional): Your own metadata stored with the job. Example: {"meeting_id": "m-123"}
- `custom_spelling` (`boolean`, optional): Apply the spelling rules in custom_spelling_config (alpha). Example: false.
- `custom_spelling_config` (`object`, optional): Spelling rules: spelling_dictionary maps the wanted spelling to the variants to replace. Example: {"spelling_dictionary": {"SQL": ["Sequel"]}}
- `custom_vocabulary` (`boolean`, optional): Use the custom vocabulary in custom_vocabulary_config (beta). Example: false.
- `custom_vocabulary_config` (`object`, optional): Custom vocabulary: vocabulary is a list of words or objects with value and intensity. Example: {"vocabulary": ["Scalekit", {"value": "Gladia", "intensity": 0.5}]}
- `diarization` (`boolean`, optional): Identify speakers. Example: true.
- `diarization_config` (`object`, optional): Speaker settings when diarization is on: number_of_speakers, or min_speakers and max_speakers. Example: {"min_speakers": 1, "max_speakers": 3}
- `language_config` (`object`, optional): Languages: languages is a list of language codes (one fixes the language, none auto-detects) and code_switching true detects language per utterance. Example: {"languages": ["en"], "code_switching": false}
- `model` (`string`, optional): Transcription model. solaria-1 (default) covers 100+ languages; solaria-3 is most accurate for English, French, German, Spanish and Italian (one language only). Example: solaria-1. One of: `solaria-1`, `solaria-3`, `solaria-fusion`.
- `named_entity_recognition` (`boolean`, optional): Detect named entities (alpha). Example: false.
- `pii_redaction` (`boolean`, optional): Redact personal data in the transcript. Example: false.
- `pii_redaction_config` (`object`, optional): Redaction settings: entity_types lists presets (such as GDPR or HIPAA_SAFE_HARBOR) or entity types to redact; empty redacts the default set. Example: {"entity_types": ["GDPR"]}
- `punctuation_enhanced` (`boolean`, optional): Use enhanced punctuation (alpha). Example: false.
- `sentences` (`boolean`, optional): Return the transcript split into sentences. Example: true.
- `sentiment_analysis` (`boolean`, optional): Detect sentiment and emotion per sentence. Example: false.
- `subtitles` (`boolean`, optional): Generate subtitles. Example: false.
- `subtitles_config` (`object`, optional): Subtitle settings: formats (srt, vtt) and optional duration and character limits. Example: {"formats": ["srt", "vtt"]}
- `summarization` (`boolean`, optional): Generate a summary. Example: true.
- `summarization_config` (`object`, optional): Summary settings: type is general (default), bullet_points or concise. Example: {"type": "bullet_points"}
- `translation` (`boolean`, optional): Translate the transcript. Example: false.
- `translation_config` (`object`, optional): Translation settings when translation is on: target_languages (list of codes) and optional model and match_original_utterances. Example: {"target_languages": ["fr", "de"]}

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

### `gladia_live_session_delete`

Delete Live Transcription · Destructive

Delete a live transcription session with its recorded audio and transcript.
Returns no content on success.
Use this to remove data you no longer need.

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

Inputs:

- `job_id` (`string`, required): ID of the live transcription job, returned when it was created or by the list tool. Example: 45463597-20b7-4af7-b3b3-f5fb778203ab.

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

### `gladia_transcription_delete`

Delete Transcription · Destructive

Delete a pre-recorded transcription job with its audio file and transcript.
Returns no content (202 or 204) on success.
Use this to remove data you no longer need.

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

Inputs:

- `job_id` (`string`, required): ID of the pre-recorded transcription job, returned when it was created or by the list tool. Example: 45463597-20b7-4af7-b3b3-f5fb778203ab.

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

## Related

Other Transcription connectors ([all 35](/agentkit/connectors/?category=transcription)):

- [Chorus](/agentkit/connectors/chorus/): Scalekit connector, Username and password, 9 tools
- [Diarize](/agentkit/connectors/diarize/): Scalekit connector, Access token, 3 tools
- [Gong](/agentkit/connectors/gong/): Scalekit connector, OAuth, 69 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 |
