◎ Discry Score
firecrawl.com
ai · API
A
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 99 (A)
Discry your API →
AI · RANK #1 OF 43

Firecrawl is Agent-Ready to agents.

Discry independently scored how well an AI agent can discover and understand the Firecrawl API from what’s public — not whether it’s usable. Below: every signal we checked, what’s costing the score, and what to change.

Discry your API — freeView the docs ↗

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 · 98/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Pass
llms.txtAn llms.txt index that points agents to the docs that matter.Pass
llms.txt qualityThe llms.txt is focused, current, and well under the size budget.Pass
llms-full.txtA full-text bundle agents can load in one request.Pass
AGENTS.mdAn AGENTS.md that tells coding agents how to build on the API.Pass
.well-known/mcp.jsonA discoverable MCP manifest at a well-known path.Partial
MCP registryThe API is listed in a public MCP registry.Pass
robots.txt AI directivesrobots.txt allows (or explicitly guides) AI crawlers.Pass
SitemapA sitemap so agents can enumerate the docs surface.Pass

Comprehension

55% of score · 100/100
Task-oriented descriptionsEndpoints described by what they accomplish, not just their shape.Pass
Realistic examplesRunnable, real-world request/response examples.Pass
Multi-step workflowsDocs that chain calls into complete jobs an agent can follow.Pass
Error-recovery guidanceDocumented failure modes and how to recover from them.Pass
Answer-first formatThe answer leads; preamble does not bury it.Pass
Capability boundariesClear limits — what the API can and cannot do.Pass
Naming consistencyConsistent, predictable naming across endpoints.Pass
Heading hierarchyClean heading structure agents can navigate.Pass
Markdown docsDocs available as clean markdown, not JS-rendered HTML only.Pass
Token efficiencyDocs are concise enough to fit an agent context window.Pass

What we found

  • An agent integrating Firecrawl finds near-perfect discovery: llms.txt + a 712KB llms-full.txt, AGENTS.md and CLAUDE.md in the primary repo, public OpenAPI specs (openapi.json + v1-openapi.json), and the official MCP server listed across PulseMCP, Glama, and Smithery — it is discoverable through virtually every channel an agent checks.
  • Firecrawl ships a purpose-built agent onboarding path: a keyless free tier, WorkOS ID-JAG agent auth, an /ai-onboarding 'Build with AI' page, and installable agent skills for Claude Code/OpenCode — the docs are explicitly written for autonomous agents, not just humans.
  • The Errors page is best-in-class for self-correction: every error maps to cause -> remedy -> retryable, with a copy-pasteable backoff snippet and explicit Retry-After guidance, so an agent can recover from failures without trial-and-error.
  • Comprehension is essentially friction-free: every doc page is available as clean markdown (.md), examples use realistic values across Python/Node/cURL/PHP, endpoint descriptions are task-oriented, and rate/concurrency boundaries are fully documented.
  • The one soft spot: /.well-known/mcp.json exists but points at a staging Mintlify preview host (firecrawl.main-kill-isr.mintlify.me) and declares only a server endpoint with no tool list — an agent relying on that file alone could hit a non-canonical endpoint.

What to change

Prioritized by impact on discoverability. You (or your docs platform) deploy these — Discry never touches your API.

  1. 01Fix /.well-known/mcp.json to point at the canonical production MCP endpoint (not the main-kill-isr.mintlify.me staging host) and include tool declarations so agents can introspect available tools from the discovery file alone.
  2. 02Document idempotency support — if scrape/crawl/batch endpoints accept an idempotency key (or explain why retries are safe), state it explicitly so agents can retry network failures without creating duplicate jobs.
  3. 03Close the remaining gaps in the Errors catalog (currently flagged 'non-exhaustive') so agents can recover from every documented failure mode without having to open an issue.

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.

API key (Authorization: Bearer fc-...)WorkOS ID-JAG agent authKeyless free tier (no API key, rate-limited fallback)✓ Error format documented✓ Rate limits documented✓ Pagination documented✕ Idempotency documented

Execution documentation is comprehensive and machine-parseable: a uniform JSON error shape ({success:false, error, details}), a dedicated Errors catalog mapping each code to cause/remedy/retryable, per-plan rate and concurrency limits with explicit Retry-After honoring, and cursor-based pagination on listing endpoints. Idempotency keys are not documented (the only stateful-retry mechanism is reusable browser profiles).

See your own Discry Score.

Drop your API docs URL. See what an agent sees — in 60 seconds, free.

Discry your API — free