Connect your AI assistant to the Greenlight Audio catalogue.
The Greenlight MCP server exposes our audiobook catalogue to Claude, ChatGPT, Cursor, and any other client that speaks the Model Context Protocol. It runs remotely — nothing to install — over MCP Streamable HTTP with stateless JSON. The server name is greenlight.
https://www.greenlight.film/api/mcpBrowse & search the catalogue
Browse the 450+ audiobooks on greenlight-audio.com, filtered by language or genre. Returns metadata cards with cover art and short preview URLs. The cross-catalogue search and fetch tools are free too.
list_audiobooksChapter lists & durations
Full metadata for one audiobook by slug — narrator voice plus every chapter with its duration, so an assistant can plan a listening session before fetching any audio.
get_audiobookStreamable chapter audio
A streamable audio URL for any single chapter — the building block for AI-driven listening apps, playlist builders, and voice agents that queue up real narration.
get_audiobook_chapter_audioThe same endpoint also serves the greenlight.film development slate and the greenlight-publishing.com book catalogue — 18 read-only tools in total. See the full tool reference on greenlight.film.
Claude Code
One command from your terminal:
claude mcp add --transport http greenlight \ https://www.greenlight.film/api/mcp
claude.ai
Settings → Connectors → Add custom connector, then paste the endpoint URL. Claude walks you through the OAuth sign-in; approve access and you’re connected.
https://www.greenlight.film/api/mcp
ChatGPT
Settings → Connectors → add the endpoint URL with OAuth authentication. The server implements the search + fetch contract, so it works with deep research out of the box.
https://www.greenlight.film/api/mcp
VS Code / Cursor
Add an HTTP MCP server with the same URL — e.g. in Cursor’s .cursor/mcp.json with an API key:
{
"mcpServers": {
"greenlight": {
"url": "https://www.greenlight.film/api/mcp",
"headers": {
"Authorization": "Bearer gl_pk_your_key_here"
}
}
}
}OAuth 2.1 + PKCE
Hosts like Claude and ChatGPT discover the server via /.well-known/oauth-protected-resource, then prompt you to sign in with your Greenlight account and click Approve. Your access tier follows your account’s plan — no key handling needed.
API key
For Cursor, scripts, and API integrations: generate a gl_pk_... key at greenlight.film/api-docs and send it as Authorization: Bearer.
| Tier | Rate limit | Audiobook access |
|---|---|---|
| Free | 10 req/hour | Browse and search audiobooks (list_audiobooks, search, fetch, get_catalog_stats) |
| Starter | 100 req/hour | + Audiobook detail with chapter lists and durations (get_audiobook) |
| Growth | 1,000 req/hour | + Streamable chapter audio URLs (get_audiobook_chapter_audio) |
| Enterprise | 10,000 req/hour | Everything above, at production volume |
Requests are metered against your hourly rate limit; get_license_status is free and unmetered — call it first to see your tier and remaining quota. Plans and pricing live at greenlight.film/pricing.
“Find a French-language Greenlight audiobook and queue up chapter one.”
list_audiobooks{ "language": "FR" }FreeThe assistant filters the catalogue by language and gets back metadata cards — titles, slugs, genres, covers, and short preview URLs.
get_audiobook{ "slug": "the-chosen-title" }StarterIt pulls the full detail for the pick: narrator voice and the complete chapter list with durations — enough to know how long chapter one runs.
get_audiobook_chapter_audio{ "slug": "the-chosen-title", "chapter_number": 1 }GrowthFinally it fetches the streamable audio URL for chapter one and hands it to your player.
Every audiobook also has a public page at greenlight-audio.com/catalogue/{slug} that assistants can cite, and cross-catalogue fetch ids look like audiobook:{slug}.
Rather listen the human way?
The whole catalogue is right here on the site — first chapters free on every title.