Quickstart
Audit a documentation site and read its score in three steps.
1. Start an audit
RUN=$(curl -s -X POST 'https://agentfit.dev/audit?async=true' \
-H 'Content-Type: application/json' \
-d '{"base_url":"https://docs.stripe.com"}' | jq -r .run_id)
echo "$RUN"
Python
import requests
resp = requests.post("https://agentfit.dev/audit", params={"async": "true"},
json={"base_url": "https://docs.stripe.com"}, timeout=10)
run_id = resp.json()["run_id"]
JavaScript
const r = await fetch("https://agentfit.dev/audit?async=true", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ base_url: "https://docs.stripe.com" }),
});
const runId = (await r.json()).run_id;
2. Poll until done
curl -s "https://agentfit.dev/api/public/audit/$RUN" | jq '.status, .total_score'
Poll every two to three seconds until status is done (or error). A typical audit
finishes in 10–30 seconds.
3. Read the report
When status is done, the report is nested under report, with total_score and a
per-criterion breakdown. The human-readable share page lives at
https://agentfit.dev/r/$RUN.
Use the CLI instead
agentfit audit https://docs.stripe.com --format both
Import into Postman
Prefer a GUI? Import the AgentFit collection and run every endpoint:
In Postman, choose Import → Link and paste
https://docs.agentfit.dev/agentfit.postman_collection.json, or download the file and import it. The
collection includes the audit, poll, history, diff, and browse requests with a ready-to-edit
baseUrl variable.
Next steps
- Start an audit — full request reference.
- Authentication and rate limits.
- Connect from an agent (MCP).