◎ Discry Score
cash flow frog.com
analytics · API
D
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 92 (A)
Discry your API →
ANALYTICS · RANK #22 OF 29

Cash Flow Frog is Poor to agents.

Discry independently scored how well an AI agent can discover and understand the Cash Flow Frog 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 · 43/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Fail
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.Partial
llms-full.txtA full-text bundle agents can load in one request.Fail
AGENTS.mdAn AGENTS.md that tells coding agents how to build on the API.Fail
.well-known/mcp.jsonA discoverable MCP manifest at a well-known path.Fail
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.Partial

Comprehension

55% of score · 59/100
Task-oriented descriptionsEndpoints described by what they accomplish, not just their shape.Partial
Realistic examplesRunnable, real-world request/response examples.Fail
Multi-step workflowsDocs that chain calls into complete jobs an agent can follow.Partial
Error-recovery guidanceDocumented failure modes and how to recover from them.Fail
Answer-first formatThe answer leads; preamble does not bury it.Pass
Capability boundariesClear limits — what the API can and cannot do.Partial
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 can actually act on Cash Flow Frog today: a first-party MCP connector (mcp.cashflowfrog.com/mcp) is documented in the help center and listed on PulseMCP and Zapier MCP, letting an agent list companies, pull live forecasts, and create/update projections through natural language with no glue code.
  • The discovery front door is agent-friendly: robots.txt explicitly welcomes GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, and Google-Extended, and a well-structured llms.txt maps the whole product, so an agent researching the tool will find it and understand what it does.
  • There is no machine-readable API surface: an agent looking for an OpenAPI spec, an endpoint reference, or a .well-known/mcp.json finds nothing (all 404), so it cannot discover the connector's exact tool names, parameter types, or return shapes before calling.
  • Examples are conversational, not code: the connector docs give realistic natural-language prompts ("What will our bank balance be on June 30?") but zero request/response samples or per-tool schemas, so an integrating agent has nothing copy-pasteable to reason from.
  • No error-recovery or limits documentation: an agent that hits a failed projection, an auth error, or a rate limit has no documented codes or recovery steps — it would only learn boundaries by failing.

What to change

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

  1. 01Publish a .well-known/mcp.json (or a linked tool manifest) declaring the connector's tools, parameters, and auth flow — this is the single highest-impact discovery fix and matches exactly how agents look for MCP servers today.
  2. 02Add a schema/OpenAPI-style reference for the connector operations (list companies, get forecast, check bank balance, create/update/delete projection, build/rename scenario) with parameter types and example request/response payloads.
  3. 03Document error codes and recovery steps for the connector (auth/sign-in failure, invalid date range, projection conflicts, rate limits) so an agent can self-correct instead of failing blind.
  4. 04Document capability boundaries and rate limits for the AI connector explicitly — what it cannot do, sync latency, and per-request limits — rather than only 'this list will keep growing.'
  5. 05Surface the agent-facing docs in discovery: add an 'AI Connector / MCP' section to llms.txt and the sitemap pointing at mcp.cashflowfrog.com and the connector help articles, so agents find the connector docs instead of only marketing pages.

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.

OAuth/credential sign-in (MCP connector)API key (via Zapier MCP)✕ Error format documented✕ Rate limits documented✕ Pagination documented✕ Idempotency documented

The only documented programmatic surface is a first-party MCP connector at mcp.cashflowfrog.com/mcp, which authenticates via Cash Flow Frog credential sign-in (API-key access is available through the Zapier MCP wrapper). No REST endpoints, error response format, rate limits, pagination, or idempotency behavior are documented anywhere.

See your own Discry Score.

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

Discry your API — free