Agent integration

Resolve, preflight, act, report.

CrawlDex gives web agents a compact operating loop for deciding whether to attempt a task on a site and for reporting what actually happened.

Loop 5 steps resolve -> inspect -> preflight -> act -> report

Reports opt-in redacted outcome evidence

Trust score-neutral until reporter verification

Before acting

Preflight a task

POST /api/v1/preflight

curl -sS -X POST https://crawldex.com/api/v1/preflight \
  -H "content-type: application/json" \
  -d '{
    "site": "example.com",
    "task": "subscriptions.cancel",
    "agent_profile": {
      "stack": "browser-use",
      "model": "gpt-5.5",
      "browser_runtime": "chromium"
    }
  }'

Use the recommendation to decide whether to proceed autonomously, keep the user present, choose an alternative, or collect fresher evidence.

After acting

Report the outcome

POST /api/v1/runs

curl -sS -X POST https://crawldex.com/api/v1/runs \
  -H "content-type: application/json" \
  -H "x-crawldex-agent-key: $CRAWLDEX_AGENT_KEY" \
  -d '{
    "site": "example.com",
    "task": "subscriptions.cancel",
    "outcome": "success_with_handoff",
    "friction": ["login_required"],
    "steps": 18,
    "duration_sec": 142,
    "evidence": {
      "hash": "sha256:REDACTED",
      "artifact_types": ["redacted_trace"]
    }
  }'

Do not submit secrets, cookies, private screenshots, payment data, or unredacted user content.

GET /api/v1/resolve?q=...