AI tools for integrators
Point your AI coding assistant (Cursor, Claude Code, ChatGPT) at our docs and let it build the integration. We expose a machine-readable OpenAPI 3.1 spec, an llms.txt index, a drop-in agent skill, and a ready MCP config — plus copy-paste prompts for the most common tasks.
Vibe-coding = you describe what you want in plain language and the AI writes the code. Copy a prompt below, paste it into your AI assistant, and follow along. If something fails, paste the exact error back and ask it to fix — it usually takes a few iterations.
- Get your API key in the dashboard (Settings → API Keys).
- Copy the Quickstart prompt below into Cursor / Claude / ChatGPT.
- Paste your key when asked (keep it on the server, never in a browser).
- Ask for the next task (start a stream, upload a video, …) using the other prompts.
Point your AI here
Connect via MCP
Add this to your MCP client config (Claude Desktop claude_desktop_config.json, Cursor .cursor/mcp.json). It exposes every endpoint as an MCP tool by pointing an OpenAPI→MCP bridge at our spec — replace pk_your_key_here with your key. Any OpenAPI-MCP bridge works.
{
"mcpServers": {
"dcast-partner-api": {
"command": "npx",
"args": ["-y", "openapi-mcp-server", "https://api.dcast.pro/api/v1/docs/openapi.json"],
"env": {
"API_BASE_URL": "https://api.dcast.pro/api/v1",
"API_BEARER_TOKEN": "pk_your_key_here"
}
}
}
}Ready-made prompts
Copy a prompt into your AI assistant. Each one already references our docs so the assistant has the context it needs.
I'm integrating the DCAST Partner API. Read the machine-readable docs here: - Index: https://api.dcast.pro/api/v1/docs/llms.txt - OpenAPI 3.1 spec: https://api.dcast.pro/api/v1/docs/openapi.json Auth is a Bearer key in the header: "Authorization: Bearer pk_...". Generate a typed HTTP client from the OpenAPI spec, read the key from an env var (never hardcode it), and write a function that verifies the key by calling GET https://api.dcast.pro/api/v1/me and printing the result.
Using the DCAST Partner API (https://api.dcast.pro/api/v1/docs/openapi.json, Bearer pk_ key), write a function that creates a live stream via POST https://api.dcast.pro/api/v1/streams with an Idempotency-Key header (UUID), then returns the ingest (RTMP/SRT) URL and the playback URL from the response. Handle 429 with exponential backoff and treat 409 as an idempotent replay.
Using the DCAST Partner API (https://api.dcast.pro/api/v1/docs/openapi.json), implement the 2-step video upload: 1) POST https://api.dcast.pro/api/v1/videos/upload to get an upload token/URL, 2) PUT the file bytes to it, 3) poll GET https://api.dcast.pro/api/v1/videos/:id until processing is complete. Show progress and handle errors from the { ok, data, error } envelope.Using the DCAST Partner API (https://api.dcast.pro/api/v1/docs/openapi.json), create a restream via POST https://api.dcast.pro/api/v1/restreams and add multiple destinations (e.g. YouTube + Twitch) to fan one ingest out to all of them. Return the ingest URL and the per-destination status.
Using the DCAST Partner API (https://api.dcast.pro/api/v1/docs/openapi.json), create a paid subscription tier via POST https://api.dcast.pro/api/v1/tiers, then create a checkout for a buyer via POST https://api.dcast.pro/api/v1/checkout (send an Idempotency-Key). Explain how the buyer completes payment and how I get notified.
Using the DCAST Partner API (https://api.dcast.pro/api/v1/docs/openapi.json), register a webhook endpoint via POST https://api.dcast.pro/api/v1/me/webhooks, then write the receiver: verify the HMAC signature header on every delivery BEFORE trusting the payload, respond 2xx fast, and process asynchronously.