canonical: https://jentic.com/apis/hedyai.com/hedyai

# Hedyai Hedy API

Jentic publishes the only available OpenAPI specification for Hedy API, keeping it validated and agent-ready. The Hedy AI meeting assistant exposes recorded meeting sessions and AI-extracted highlights through this API. Across four read-only endpoints, agents can list and inspect sessions, retrieve transcripts and summaries, and pull individual highlights such as decisions, action items, and questions. Authentication uses an API key in the X-API-Key header; this is the hedyai.com canonical surface for the same Hedy product.

## For AI agents

List Hedy AI meeting sessions and retrieve their transcripts, summaries, and AI-generated highlights across 4 read-only endpoints.

## Scope

Does not record audio, transcribe live calls, or create sessions - use for read-only access to existing Hedy sessions and highlights only.

## Capabilities

- List recent Hedy meeting sessions, paginated and filterable by date, via GET /sessions
- Retrieve a single session's transcript and summary by id via GET `/sessions/{sessionId}`
- List AI-generated highlights such as decisions, action items, and questions via GET /highlights
- Fetch the full text and source-session reference of one highlight via GET `/highlights/{highlightId}`
- Filter highlights to a specific session or category for downstream summarization workflows

## Use cases

### Weekly Status Email Generation

An assistant compiles a weekly status email by pulling the past week's Hedy highlights and grouping them into decisions, action items, and open questions. The /highlights endpoint returns highlights with their source session reference, so the assistant can attribute each item to the meeting it came from.

Example prompt: Call GET /highlights filtered to the last 7 days and group the results into decisions, action items, and questions

### CRM Meeting Note Sync

A sales operations workflow syncs Hedy meeting summaries into CRM contact records. For each new session, the workflow calls `/sessions/{sessionId}` to fetch the summary and key highlights, then writes them onto the matched contact. No transcription or summarisation is needed locally - Hedy delivers the structured output.

Example prompt: For each session in GET /sessions from the last 24 hours, retrieve the summary via GET `/sessions/{sessionId}` and append it to the matching CRM contact

### Searchable Meeting Memory

A team builds an internal search tool over their meeting history by indexing Hedy sessions and highlights into a vector store. The /sessions and /highlights endpoints supply the raw text content to embed; the agent only handles the index and retrieval layer.

Example prompt: Iterate GET /sessions, fetch each session's transcript via GET `/sessions/{sessionId}`, and embed the text into a vector store

### AI Agent Meeting Assistant

A Claude or GPT assistant answers 'what did we decide in yesterday's meeting?' by calling Hedy through Jentic. The agent searches by intent, loads the /highlights operation, filters by date and category, and returns a concise answer with citations to specific session ids.

Example prompt: Search Jentic for 'list meeting highlights', load the operation, filter to yesterday's date, and return the action items with their source session ids

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/sessions` | List meeting sessions |
| GET | `/sessions/{sessionId}` | Get session transcript and summary |
| GET | `/highlights` | List AI-generated highlights across sessions |
| GET | `/highlights/{highlightId}` | Get a single highlight's content and source session |

## Key resources

- **Sessions** — List meeting sessions and retrieve full transcripts and summaries by session id
- **Highlights** — List and fetch AI-extracted highlights such as decisions, action items, and questions

## Why Jentic

- **Setup:** Wiring Hedy by hand means passing your API key as a bearer value in the Authorization header on every read call. Through Jentic you install once, import the Hedy API from the API Directory, store the key once, and your agent calls it.
- **Permission scoping:** Hedy puts the resource id in the URL path (`/sessions/{sessionId}`, `/highlights/{highlightId}`), so a rule can pin your agent to one session: it can read that session and its highlights and nothing else. The API is read-only, and you choose the operations it may call, so anything beyond those reads is not included unless you add it.
- **Credential handling:** Your Hedy X-API-Key is stored once, encrypted, by your own Jentic One instance and injected in the Authorization header at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'list meeting highlights' or 'get a meeting session', and Jentic returns the matching Hedy operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Fireflies.ai API** — Fireflies is a competing AI meeting assistant offering similar transcript and highlight extraction
- **Grain API** — Grain captures meeting recordings and highlights and is a direct alternative to Hedy
- **Gong API** — Gong focuses on revenue intelligence over recorded calls, complementary to Hedy's general meeting capture

## FAQ

### Why is there no official OpenAPI spec for Hedy API?

Hedy AI does not publish an OpenAPI specification for direct download. Jentic generates and maintains this spec so that AI agents and developers can call Hedy via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the Hedy API use?

Hedy uses an API key passed in the X-API-Key header (the ApiKeyAuth scheme in the spec). Through Jentic, the key is stored encrypted in the vault and the agent receives scoped access at execution time.

### Can I retrieve a meeting transcript with this API?

Yes. GET `/sessions/{sessionId}` returns the session record including the transcript and summary fields, so an agent can pull the full meeting text in a single call.

### What are the rate limits for the Hedy API?

The OpenAPI spec does not declare hard rate limits; Hedy applies fair-use limits per workspace. For high-volume sync workflows, paginate /sessions by date and cache transcripts locally rather than re-fetching.

### How do I list recent meeting highlights through Jentic?

Run pip install jentic, have the agent search for 'list meeting highlights', load the GET /highlights operation, and execute it filtered to the desired date range.

### Does the Hedy API support creating or updating sessions?

No. The current API surface is read-only - agents can list and retrieve sessions and highlights, but session creation happens inside the Hedy app via the meeting recorder.

### Can I limit what my agent is allowed to do with the Hedy API?

Yes. Because you run Jentic One yourself, your own rules decide which Hedy operations and credentials the agent may use, and the Hedy API is read-only to begin with, so an agent can only list and read sessions, transcripts, summaries, and highlights. Since Hedy puts the resource id in the URL path (GET `/sessions/{sessionId}` and GET `/highlights/{highlightId}`), you can pin the agent to a single session so it reads only that session and its highlights and nothing else. Any operation you do not grant, such as GET /sessions or GET /highlights, stays out of reach until you explicitly allow it.
