Reference

API Reference

REST endpoints exposed by the webhook/API server, plus the tool surface available over MCP. All incident-queueing endpoints return immediately with status: queued — a background worker does the actual diagnosis.

Health & audit

GET/healthAgent & queue worker status
GET/audit?org_id=Org-scoped incident audit log
GET/metricsPrometheus metrics (incidents, tool calls, approvals)

Webhooks

POST/webhook/manualManually queue an incident
POST/webhook/alertmanagerPrometheus Alertmanager alerts
POST/webhook/githubGitHub Actions workflow failures

$ curl -X POST http://localhost:8000/webhook/manual \

  -H "Content-Type: application/json" -H "X-Org-ID: acme-corp" \

  -d '{ "type": "k8s", "namespace": "default", "pod": "myapp-pod", "description": "CrashLoopBackOff test" }'

{ "status": "queued", "incident_id": "INC-2026-004" }

Org config (BYOK)

PUT/orgs/{org}/configStore org-owned API keys
GET/orgs/{org}/config/statusWhich credentials are configured (values never returned)

$ curl -X PUT http://localhost:8000/orgs/acme-corp/config \

  -H "Content-Type: application/json" \

  -d '{ "credentials": { "ANTHROPIC_API_KEY": "sk-ant-...", "SLACK_WEBHOOK_URL": "https://hooks.slack.com/...", "GITHUB_TOKEN": "ghp_..." } }'

Org documentation

POST/orgs/{org}/docsUpload runbooks & policies
GET/orgs/{org}/docsList org documentation

MCP tools (32 total)

DevOps action tools (get_k8s_context, run_kubectl, get_github_logs, sync_argocd_app, helm_rollback, get_cloud_resource) mirror the built-in agent's capabilities. Platform tools manage the multi-tenant layer:

configure_org_credentialsStore org keys for shared webhook/API deployments
get_org_config_statusCheck which keys are set (values never returned)
enqueue_incidentQueue an incident for the background worker
diagnose_incidentRun the full Claude agent loop with the org's Anthropic key
get_incident_auditFetch org-scoped audit history
list_org_docs / get_org_docRead runbooks and policies