gitrakz

name	gitrakz
description	Drive a self-hosted gitrakz instance — the tool that syncs a GitHub user's activity into local SQLite, renders a filterable timeline and derived work sessions, and runs deterministic programmatic templates that export to CSV/PDF/JSON. Install or run it with Docker, then query its bearer-protected REST API under /api/v1 (owners, repos, timeline, sessions, sync + sync status, templates CRUD, LLM template generation, run, export) — or drive the same capabilities as MCP tools over streamable HTTP (/mcp) or stdio (`gitrakz mcp`). Use when the user wants to set up gitrakz, trigger or check a GitHub activity sync, pull a timeline or work-sessions timesheet, or run and export a template.
homepage	https://github.com/psyb0t/gitrakz
user-invocable	true
metadata	{"openclaw":{"emoji":"🛰️","primaryEnv":"GITRAKZ_URL","requires":{"bins":"[Truncated]"}}}
permissions	{"network":"Outbound HTTP only to the user-configured GITRAKZ_URL (the local gitrakz server). gitrakz itself shells out to the GitHub CLI and reaches api.github.com to sync activity; point it only at a gitrakz instance the user runs.","shell":"bash, curl, and the docker / gitrakz-wrapper commands shown in references/setup.md, for user-requested setup or verification. gitrakz reads its GitHub token from `gh auth token`.","filesystem":"Normal use reads GITRAKZ_URL and GITRAKZ_AUTH_TOKEN from the environment. Setup writes only the owner-only ~/.config/gitrakz/.env and docker-compose.yml."}

gitrakz

gitrakz is a self-hosted GitHub activity tracker. It syncs a user's gh activity into a local SQLite database, renders a filterable timeline and derived work sessions, and runs programmatic templates — deterministic transform pipelines over the timeline that export to CSV, PDF, or JSON. One Go binary with the Svelte UI embedded; no external services required.

For setup — the installer, the direct-Docker path, config, and auth — read references/setup.md before touching a stack.

Security and safety

• This skill drives a gitrakz instance the user already runs. Take GITRAKZ_URL (e.g. http://127.0.0.1:8080) and, if set, GITRAKZ_AUTH_TOKEN from the environment or ask — do not hunt the workspace for tokens.
• gitrakz authenticates to GitHub with a token from gh auth token, injected at runtime and never written to disk. It reads activity only; mount a read-scoped token where you control the scope.
• POST /api/v1/sync starts a GitHub sync (network + rate-limit spend) and POST /api/v1/run may call an LLM if the template uses describe-work or a prose block. Only trigger these for the task the user named.
• GITRAKZ_ELELEM_BASE_URL receives commit titles / diffs when LLM steps run. Point it only at endpoints the user trusts. Template output is typed display blocks, never author-supplied HTML.
• gitrakz also exposes its capabilities as MCP tools (see "MCP" below), over the same two surfaces: streamable HTTP at /mcp (Bearer-gated exactly like /api/v1 when GITRAKZ_AUTH_TOKEN is set) and stdio via the binary's mcp subcommand. Every MCP tool wraps the same read-mostly service layer as the REST API above — same sync/LLM cost caveats apply to gitrakz_trigger_sync and gitrakz_run_template.

Use it for

• Installing or running gitrakz and confirming it is up.
• Triggering an incremental sync (POST /api/v1/sync) and checking progress (GET /api/v1/sync/status).
• Pulling a filtered timeline or a derived work-sessions timesheet.
• Listing, creating, editing, and running templates, and exporting a run to CSV / PDF / JSON.
• Any of the above through MCP tools instead of raw REST calls, when the driving client speaks MCP (see "MCP" below).

Do not use it for