PropLine developer resources
PropLine is a REST API for sports betting odds across 54 sports and 25 books, with every player prop graded against the official box score. Base URL https://api.prop-line.com, auth with ?apiKey= or an X-API-Key header. A free key is 1,000 requests a day with no credit card.
curl "https://api.prop-line.com/v1/sports?apiKey=YOUR_KEY"
curl "https://api.prop-line.com/v1/sports/baseball_mlb/events/12649/odds?apiKey=YOUR_KEY&markets=batter_hits"Reference
What the API does and how to call it.
- API reference
Every endpoint, parameter, response schema, market key and sport key, with curl examples.
- OpenAPI 3.1 spec
Machine-readable description of the whole surface — import it into Postman, Insomnia, an OpenAPI client generator, or an agent that builds calls from a schema.
- Interactive API docs
Swagger UI served by the API itself, always matching the deployed build.
- Recipes
Working code for real questions: find +EV plays, grade closing-line value, scan for arbitrage, backtest a prop, push line moves to Discord.
- Versioning & deprecation policy
What can change inside /v1, what ships as /v2, and the 12-month sunset window signalled by Deprecation and Sunset headers.
Clients
Published on the standard registries — install one line, skip the HTTP plumbing.
- Python SDK — propline
Typed client covering every endpoint, plus HMAC webhook signature verification.
pip install propline - Node / TypeScript SDK — propline
Dual ESM + CJS build with full type definitions, zero runtime dependencies, Node 18+.
npm install propline - CLI — propline-cli
Query the API from a terminal or a shell script: propline live, propline odds mlb, propline ev mlb 12345 --json.
npm install -g propline-cli - MCP server — propline-mcp
Model Context Protocol server exposing the API as tools for Claude, Cursor, ChatGPT and other agents. Hosted Streamable HTTP endpoint: https://mcp.prop-line.com/mcp
npx -y propline-mcp
For AI agents
Machine-readable documentation and native tool access.
- llms.txt
Index of the API with a when-to-use section — the file to hand an agent that has to decide whether PropLine answers its question.
- llms-full.txt
The complete documentation inlined as one markdown file. Fetch this single URL to give a model everything it needs to write working PropLine code.
- Markdown content negotiation
Send Accept: text/markdown to any page on this site and get markdown back instead of HTML (acceptmarkdown.com).
- Built for agents
MCP setup, structured machine-actionable errors, and why graded prop results are the ground truth an agent can check itself against.
Operations
Push delivery, live data-health signals, and worked examples.
- Webhooks
Push delivery for line_movement, resolution, steam and market_suspended events, HMAC-signed with retries and a delivery log.
- Freshness
Per-book staleness, live, no API key required — poll it to decide whether a book is worth reading.
- Status
Externally monitored uptime and incident history.
- Open-source examples
Reference implementations: arbitrage finder, CLV tracker, parlay EV calculator.