Portable, project-agnostic operating kit for Claude Code and Cursor. A disciplined builder and strategic-vetting persona, playbooks, subagent templates, and memory seeds. Installs with one prompt.
This repository also configures its own agents. See what operating-kit tells them →
Agent Claude Code
Comprehensive pre-ship review of all changes since the last deploy or a specified commit. Walks correctness, atomicity/races, error handling, data-store hygiene, security, type safety, tests, integration, performance, observability, and any project-specific invariants. Use after any sprint and always before deploying.
Agent Claude Code
Runs tests, builds, deploys to production, verifies live, and updates the state doc with the confirmed revision. Use before any production deploy. Stops if tests fail; never ships broken code.
Agent Claude Code
Pulls recent production logs filtered for errors, warnings, and anomalies, a quick health check after any deploy, after a load test, or any time you want to know if something is going wrong. Treats logs as the only acceptable primary source for incident analysis.
Agent Claude Code
Run at the end of every significant work session. Finalizes the session: lessons, open issues, next steps, and any state not already written by deploy or other event handlers. Skips cleanly if nothing significant changed.
Agent Claude Code
Run at the start of every work session. Reads the canonical state doc, verifies live state, reconciles drift from external deploys or dirty shutdowns, and prints a concise briefing. Prevents stale-state mistakes.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: