claude-code-config

Triangulate plans and code from three expert viewpoints — functionality, security, and testing — across three phases (plan, implementation, review). Each iteration sharpens the position of remaining issues until all are resolved. Always use this skill when: asked to review plans, code, or branches; asked to evaluate from functionality/security/testing perspectives; asked for PR or pre-implementation review; asked to implement or develop from a plan.

Mine lessons from configured knowledge sources — review artifacts in sibling repositories, GitHub PR review comments, own session transcripts, whitelisted external references — and fold novel lessons into the skill rule set, detection hooks, and cross-skill guards, driving the change through review to a pull request. Use this skill when: a session-start notice says retrospective mining is due; asked to run retrospective mining or process due sources; asked to mine lessons from review artifacts or another repository; asked to fold accumulated lessons into skills, hooks, or rules.

Generate tests for specified code or changed code. Detects test framework automatically, generates test outlines via local LLM, then implements and verifies tests via Sonnet sub-agent. Use this skill when: asked to generate or add tests; asked to improve test coverage; asked to write tests for specific files or functions.

Run a diff review with whatever review agent is available — a local zero-cost Ollama model, Codex, or a fresh headless Claude — then return structured findings. Use this skill when: asked for a Codex review or a second-opinion review from another model; asked for a free/local/offline review; asked to review changes without spending Claude tokens; asked for an adversarial review that challenges the design/approach; asked to review uncommitted changes, a branch, or a commit with an external reviewer; asked to cross-check a triangulate review against another model.

Review changed code for reuse, quality, and efficiency, then fix any issues found. Use this skill when: asked to simplify or clean up code; asked to find duplication or refactoring opportunities; asked to improve code quality of recent changes.

Deep codebase exploration and Q&A. Searches and analyzes code structure, traces call chains, and explains how things work. Use this skill when: asked to explain how something works; asked to find all usages or callers; asked to understand codebase architecture; asked to trace data flow or control flow.

Audit Claude Code context window consumption across skills, rules, agents, hooks, CLAUDE.md, and MCP servers. Reports token overhead and prioritized savings. Use this skill when: asked about context usage; asked to audit skill/rule/MCP bloat; after adding multiple skills or rules; before adding more components.

Create a pull request with auto-generated description. Summarizes changes and generates PR body via local LLM (gpt-oss:120b), then creates the PR after user approval. Use this skill when: asked to create a PR; asked to submit changes for review; asked to open a pull request.

Audit Claude Code configuration for security issues: hardcoded secrets, overly permissive allow lists, hook injection risks, MCP supply chain, auto-run instructions. Use this skill when: setting up a new project; after modifying settings.json / CLAUDE.md / .mcp.json / hooks; before committing config changes; onboarding to a repo with existing Claude Code config.