Spotify is Good to agents.
Discry independently scored how well an AI agent can discover and understand the Spotify API from what’s public — not whether it’s usable. Below: every signal we checked, what’s costing the score, and what to change.
SCORED UNDER RUBRIC 1.2 · A full re-launch under Discry Score 2.5 — a new behavioral instrument, not comparable to these scores — is in progress.
Discovery
45% of score · 67/100Comprehension
55% of score · 88/100What we found
- Spotify provides a concise, well-structured llms.txt (668 bytes) that links directly to the OpenAPI spec and a Building with AI tutorial — exemplary agent discovery
- The OpenAPI 3.0 spec is publicly accessible at /reference/web-api/open-api-schema.yaml and explicitly allowed in robots.txt
- An agent would find multiple community MCP servers on PulseMCP and Glama for playback control and catalog search
- Documentation includes a dedicated Building with AI tutorial showing how to integrate with AI coding assistants
- Sitemap returns a 404 page-not-found HTML response, hurting standard web crawler discovery
What to change
Prioritized by impact on discoverability. You (or your docs platform) deploy these — Discry never touches your API.
- 01Add llms-full.txt with comprehensive API documentation in markdown
- 02Fix sitemap.xml to return valid XML with documentation pages listed
- 03Add AGENTS.md to the spotify GitHub org with context for coding agents
- 04Add .well-known/mcp.json pointing to community or official MCP server
- 05Include more realistic code examples with actual values across multiple languages
Execution coverage · INFORMATIONAL, UNSCORED
Whether an agent can actually complete a call and recover from errors is the deeper Audit layer — documented here, but not part of the Discry Score.
Spotify uses OAuth 2.0 (authorization code, client credentials, PKCE). JSON error responses. Rate limits documented with quota modes. Cursor-based pagination. No idempotency keys documented.