AlexisBalayre

AlexisBalayre/claude-code-power-config

Agent Claude Code

Subagents run in their own context window and return only a summary, so deep analysis never bloats the main conversation.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Use PROACTIVELY when the user asks why or how about the system architecture — service boundaries, data flow, session lifecycle, provider strategy, scaling, schema decisions, auth model, or cross-service interactions. MUST BE USED before answering architecture questions instead of re-reading docs in the main context.…

AlexisBalayre/claude-code-power-config

Agent Claude Code

Autonomously prune bad comments from code added in the current session. Dispatched by the comment-pruner Stop hook when net-new comments are detected; also usable manually for a repo-wide sweep. Deletes or rewrites violations directly in the working tree under a delete-when-uncertain policy bounded by a hard carve-out…

AlexisBalayre/claude-code-power-config

Agent Claude Code

Use PROACTIVELY to verify files follow area-specific coding standards before committing. MUST BE USED after editing three or more files under apps/, services/, or packages/, or when preparing a commit. Cross-references docs/conventions/.md for domain rules.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Use PROACTIVELY after any change to packages/acme-db/src/schema/ or newly generated migrations in packages/acme-db/src/migrations/. MUST BE USED before committing schema or migration changes. Reviews for safety, backwards compatibility, and convention compliance against docs/conventions/backend.md.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Reviews changed code for spec contradictions, protocol violations, and infrastructure anti-patterns. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Audits changed files for compliance with the project's documented rules and accepted ADR decisions. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Reviews changed code for logic and behavior defects that deterministic tooling cannot catch, including behavioral regressions on the surface tests do not cover. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Reviews the quality, accuracy, and usefulness of prose a change introduces. Strict on low-value comments and docstrings. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Reviews changed code for structural regressions that make the codebase harder to maintain. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Reviews changed code for real, reachable security issues at system boundaries. Spawned by the pr-ci-review skill.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Adversarially verifies review findings against the actual code, refuting the ones that do not hold. Spawned by the pr-ci-review skill before any finding is auto-acted or posted.

AlexisBalayre/claude-code-power-config

Agent Claude Code

Use PROACTIVELY after editing auth middleware, new API routes, WebSocket handlers, user-facing forms, secret handling, or anything touching external input validation. MUST BE USED before committing changes in sensitive areas. Reviews for injection vulnerabilities, auth/authz flaws, secrets exposure, and OWASP Top 10…

AlexisBalayre/claude-code-power-config

Hook Claude Code

Runs when the agent finishes a response, executing quality-checks.sh, convention-spot-check.sh and comment-pruner.sh (3 commands). From AlexisBalayre/claude-code-power-config.

AlexisBalayre/claude-code-power-config

Hook Claude Code

Runs before the agent uses a tool for Bash, Edit, Write and Write tool calls, executing git-safety.sh, protect-generated.sh and validate-file-naming.sh (3 commands). From AlexisBalayre/claude-code-power-config.

AlexisBalayre/claude-code-power-config

Hook Claude Code

Runs before the context is compacted, executing pre-compact-preserve.sh. From AlexisBalayre/claude-code-power-config.

AlexisBalayre/claude-code-power-config

Settings file Claude Code

Agent settings declaring 3 hook events (Stop, PreToolUse, PreCompact) and 32 allowed tools.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Analyze, decide on, challenge, and implement a PR's open review comments end to end, with a human in the loop. Use when addressing PR review feedback, responding to reviewer comments, or when the user says "address review comments".

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Generate tracker issues from merged PRs for the current sprint following strict grouping and review rules.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Ultra-compressed communication mode. Cuts token usage 75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Shared vocabulary for designing deep modules. Use when designing or improving a module's interface, deciding where a seam goes, making code more testable, or when another skill needs the deep-module vocabulary.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Create or update daily Obsidian logbook notes with issue-tracker and Git integration.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken, throwing, failing, or slow.

AlexisBalayre/claude-code-power-config

Skill Claude CodeCodex

Maintain this repo's documented language as design decisions land. Use when pinning down domain terminology, recording an architectural decision, or when another skill needs the docs kept current during a session.