How to use
- Copy the installation command and review the source package before installing it in your own agent environment.
- Check the source document for required tools, credentials and supported hosts.
- Describe your input and desired deliverable; review the output against the source material.
npx skills add https://github.com/firetiger-oss/cursor-plugin --skill firetigerBefore you start
- Installation does not include a model subscription, API allowance or permission to access your files.
- Repository stars and installation counts describe activity, not an independent quality or security test.
Rendered from the source repository. Headings, examples, code, tables, links, and referenced images are preserved.
Firetiger
Firetiger is an AI-powered observability platform. Telemetry (traces, logs, metrics) lands in Apache Iceberg tables you query with SQL, and autonomous agents monitor your services, investigate issues, and watch deployments.
This skill is a router. Identify what the user wants and invoke the matching specialized skill — each one is self-contained and carries the gotchas for its task.
How you talk to Firetiger
Two mechanisms, referenced throughout the skills:
- The Firetiger MCP server —
https://api.cloud.firetiger.com/mcp/v1(OAuth 2.0 bearer; the first tool
call opens a browser to sign in). Tools: get_ingest_credentials, get_deploy_credentials, query, monitor_pr, create_agent_with_goal, send_agent_message, read_agent_messages, resolve_url, get_subscription_status/get_checkout_url (billing), onboard_github/onboard_slack/onboard_linear (OAuth connect), and generic CRUD (schema/list/get/create/update/delete). The collections a client agent works with: agents + sessions (create/run agents), triggers + scheduled-agent-runs (automate them), investigations (AI diagnosis sessions), issues (Known Issues, IDs are FT-{n} call signs), monitoring-plans (deploy monitoring, read/delete), and connections (integrations). Always schema a collection before create/update. Some tools and collections are gated by deployment config and API-key policy.
- The `@firetiger` GitHub flow — comment
@firetigeron a pull request to have Firetiger monitor the
deployment that PR produces.
Key gotcha: if the MCP tools aren't available, the user hasn't connected the Firetiger MCP server yet. The target skill handles this — it tells the user to connect https://api.cloud.firetiger.com/mcp/v1 and sign in, then retries.
Routing
| The user wants to… | Skill | Trigger phrases |
|---|---|---|
| Onboard a project end-to-end (detect stack → instrument → connect → agent) | `firetiger-setup` | "set up Firetiger", "onboard this project", "connect my app" |
| Add OpenTelemetry instrumentation (Node/Next.js/Python/Go/Rust) | `firetiger-instrument` | "instrument my app", "add OpenTelemetry", "send traces" |
| Query traces, logs, or metrics with SQL | `firetiger-query` | "find traces", "search logs", "show me errors", "analyze latency" |
| Investigate an incident and track findings | `firetiger-investigate` | "investigate", "diagnose", "what's wrong with", "troubleshoot" |
| Monitor a PR/deployment | `firetiger-monitor-deploy` | "monitor this PR", "watch this deploy", "@firetiger" |
| Create a monitoring agent, or configure agents/triggers | `firetiger-create-agent` | "create an agent", "monitor X automatically", "schedule an agent" |
Execution
- Classify the request into one row above.
- Invoke the corresponding skill with the Skill tool (e.g.
firetiger-query). - If a request spans categories ("set up Firetiger and monitor my next PR"), handle them sequentially —
usually firetiger-setup first, then the follow-up.



