Getting Started
Developer API reference
A practical map of the backend API groups that power CloudWatcher features.
This is not a formal OpenAPI document yet. It is a developer-readable API map based on the backend route groups, useful for integration work and debugging.
API groups
- Auth API: signup, verify signup, login, login 2FA, resend 2FA, forgot password, verify forgot password, reset password, OTP send, OTP verify, restore account, profile update, 2FA setup, 2FA confirm, 2FA remove, security events, team users, and integrations.
- Cloud API: providers, connections, resources, metrics, billing, security, insights, recommendations, and logs across AWS, Azure, and GCP.
- AWS API: setup, credentials, billing, resources, metrics, logs, insights, security, alarms, default alarms, alarm metadata, credential vault, live resource actions, Lambda code view, live action safety check, optimization, and role save.
- Azure API: setup, save connection, credentials, billing, resources, metrics, logs, insights, security, alarms, default alarms, alarm metadata, and live resource actions.
- GCP API: setup, save connection, update billing, credentials, billing, resources, metrics, logs, insights, security, alarms, default alarms, alarm metadata, and live resource actions.
- Actions API: plan, plan from recommendation, preview, create, simulation log, approve, execute, rollback, history, status, savings, savings verification, audit, and registry.
- AI Keys API: save key, delete key, key status, OpenAI usage, OpenAI logs, OpenAI per-key usage, Anthropic usage, Gemini usage, and pricing.
- AI Observability API: overview, tokens, costs, models, errors, alerts, alert evaluation, budget, forecast, daily summary, weekly summary, anomalies, budget enforcement, Bedrock console, Bedrock sync, routing recommendations, prompt insights, and notification history.
- AI Ingest API: event ingest, batch event ingest, trace ingest, batch trace ingest, scoped ingest keys, and direct client SDK trace uploads.
- AI Traces API: trace list, trace detail, request detail, trace post, and batch trace post.
- Prompt and Evaluation API: prompt list, prompt create or update, prompt delete, playground run, evaluation list, and evaluation run.
- AI Quality API: feedback, comments, score views, sessions, users, and request detail workflows.
- Simulation API: create session, session status, stream session logs, terminate session, estimate cost, cached cost status, generate Terraform, start deployment, run deployment, resume deployment, cancel deployment, validate credentials, deployment status, deployment stream, PEM download, saved simulations, simulation detail, update, destroy, delete, and simulation PEM download.
- GitHub API: status, connect, disconnect, repositories, and branches.
- Resize Migration API: scope, sources, target sizes, jobs list, job create, plan create, job detail, transition, confirm classification, configure access, resume, report, explain task, and delete.
- Usage Reports and Notifications API: report preferences get, report preferences update, test report, notification settings get, notification settings update, and Slack notification test.
- VPS Logs API: agent list, create agent, patch agent config, delete agent, ingest logs, summary, clear recent logs, alarm rules, alarm create, alarm update, alarm delete, alert policy, and alert mail test.
- SaaS Admin API: system stats, tenant list, and compliance alert list for system administrators.
Authentication and roles
- Most product APIs require authentication.
- Write-heavy APIs usually require admin or operator roles.
- Some public ingest paths use webhook secret or ingest-key authentication instead of normal app login.
- Sensitive APIs such as team management, provider key management, credential vault, reports, and SaaS admin are role-gated.
Debugging pattern
- For frontend issues, identify the page, the API group it calls, and the selected provider or workspace state.
- For cloud issues, separate credential validation, provider permissions, region, quota, billing export, and unsupported capability state.
- For AI issues, check ingest key, event or trace payload, provider key, model pricing, time range, and filters.
- For simulation issues, read Terraform preview, credential validation, deployment logs, and destroy logs before retrying.
Need more help?
Keep moving with the right next guide
If your team still runs into setup issues, empty dashboards, billing delays, callback failures, or alerting problems, continue with troubleshooting before re-running the entire onboarding flow.