check_incident
Call this when a web step is blocked or failing for no clear reason (a 403 or access denied, a 429, a 5xx or a timeout, a login wall, a "verify you are human" or CAPTCHA page), when sign-in failed or an extra verification step appeared (a code, app approval, new-device check; error_class=login_auth), or when asked "is this site down?" or whether its robots rules block agents. Skip it when the page opened normally or your own tool or platform already said why it refused (e.g. robots.txt or platform policy). Look the host up once per blocked step, before deciding whether to retry. Send the public hostname, never a URL path, query or credentials; add codes only when you observed them: error_class (site_access, login_auth, human_verification, service_error), failure_stage (before_target_response, response_observed, access_denial_observed, unknown), http_status, challenge, retry_after, via. Send agent_vendor only when you know which vendor's documented robots.txt token your requests send: openai, anthropic (Claude-User), google, perplexity, meta, mistral, amazon, microsoft, apple, duckduckgo, diffbot, exa, or other if none. Browser automation usually sends a normal browser user agent, so robots.txt agent tokens may not apply; leave agent_vendor out then. The answer: summary_line, next_action (do, url, retry_after_seconds), recommendation.action, access_profile, official_paths, provider_status and observations. no_reports is not a healthy site. Then follow next_action (continue, wait, use a listed official path, hand the step to the user, stop retrying, or look up once more). On 'pending', look up once more after retry_after_seconds; do not poll. Read-only: it may fetch the host's public robots.txt, llms.txt and status feed, and the home page unless robots.txt bars it (ahn_check), never changes the target site or your data; lookups and the codes you send are counted anonymously per host and hour. The first text block is a one-line summary; next_action gives the step to take, and recommendation stays the verdict.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"site_host": "<site_host>"
},
"name": "check_incident"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| via | No | What made the failed request: http_fetch (a plain HTTP fetch), browser (browser automation) or unknown. Leave it out if unsure. | |
| test | No | Set true to mark this lookup as a test (like the header X-AHN-Test: 1): it is kept out of the anonymous counts and the answer is the same. Leave it out for a real lookup; any other value is ignored. | |
| challenge | No | header if the site's response carried a bot-challenge header, none if you saw the response and it did not. Never guess from the page body; leave it out if unknown. | |
| failed_at | No | When the step failed, as an ISO 8601 time such as 2026-11-01T07:10:00Z (UTC if no offset). Leave it out for a failure that just happened. Only used to choose the cause shown in the answer; not stored. | |
| site_host | Yes | The public hostname that is blocked or failing, lowercase, for example example.com. Only the host: no scheme, URL path, query, port or credentials. | |
| task_step | No | What you were doing, as a short lowercase code of letters, digits and underscores starting with a letter. Use status_check when you only check whether the site is up (not blocked). Leave it out otherwise. | |
| error_class | No | What stopped you, only if you observed it: site_access (page or API refused or unreachable), login_auth (sign-in failed or an extra verification step appeared), human_verification (CAPTCHA or verify-you-are-human check), service_error (error response or timeout). Leave it out if unsure. | |
| http_status | No | HTTP status the site returned: 401, 402, 403, 407, 408, 429, 451 or 500-599. Leave out if unknown. | |
| retry_after | No | Size of the Retry-After value the site sent, in seconds: none (no header, or an HTTP-date), lt_60 (under 60), 60_3600 (60 to 3600), gt_3600 (over 3600). Leave it out if you saw no response. | |
| agent_vendor | No | The vendor whose documented robots.txt token your requests send (not the model vendor): openai, anthropic (Claude-User), google, perplexity, meta, mistral, amazon, microsoft, apple, duckduckgo, diffbot, exa, or other if none of them. Leave it out if you do not know, and for browser automation that sends a normal browser user agent. | |
| failure_stage | No | How far the request got, only if you observed it: before_target_response (no answer from the site, for example a timeout or connection failure), response_observed (the site answered with an error or unexpected page), access_denial_observed (the site explicitly refused access), unknown. Leave it out if unsure. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| cause | No | ||
| links | No | ||
| summary | Yes | ||
| failures | Yes | ||
| ahn_check | No | ||
| site_host | Yes | ||
| successes | Yes | ||
| synthetic | Yes | ||
| host_facts | No | ||
| next_action | No | ||
| next_checks | No | ||
| generated_at | Yes | ||
| observations | Yes | ||
| summary_code | Yes | ||
| summary_line | No | ||
| window_hours | Yes | ||
| failure_stage | No | ||
| access_profile | No | ||
| evidence_level | Yes | ||
| feedback_token | No | ||
| official_paths | No | ||
| recommendation | No | ||
| lookup_activity | No | ||
| provider_status | No | ||
| reporting_sources | Yes | ||
| data_horizon_hours | Yes | ||
| current_site_status | Yes | ||
| official_status_page | No | ||
| provider_source_state | Yes | ||
| provider_discovery_attempted_at | No |