gitai-js

Rank the functions/modules changed this session by test-worthiness and output TEST / integration-only / SKIP per symbol. Uses graphify blast-radius when graphify-out/ exists, a static rubric otherwise. Use when you changed several units and are unsure which deserve a unit test, or before writing tests for a batch of new code.

Use when performing code review, writing or refactoring code, or discussing architecture; complements clean-code and does not replace project linter/formatter.

Use when publishing a new version of gitai to npm — cutting a release, bumping the version, shipping to the registry — or when a release was cut but the new version never reached npm, or a user reports the CLI still showing an old version.

Use when asked to generate, build, prepare, optimize or assemble a /goal command for autonomous or overnight execution — triggers: "montar goal", "gerar goal autônomo", "preparar task overnight", "goal para sessão autônoma", "otimizar goal", "goal builder".

Smoke test end-to-end do gitai CLI em repo temporário

Create, update, and manage GitHub issues using MCP tools. Use this skill when users want to create bug reports, feature requests, or task issues, update existing issues, add labels/assignees/milestones, set issue fields (dates, priority, custom fields), set issue types, manage issue workflows, link issues, add dependencies, or track blocked-by/blocking relationships. Triggers on requests like "create an issue", "file a bug", "request a feature", "update issue X", "set the priority", "set the start date", "link issues", "add dependency", "blocked by", "blocking", or any GitHub issue management task.

This skill should be used when the user mentions tickets, issues, backlog, PR, pull request, worktree, sprint, or any development project management activity. Also activate when the user says "create issue", "work on #N", "close ticket", "open PR", "merge PR", "view backlog", "create branch for issue", or when code was modified without a linked issue.

Use when bootstrapping a repo (apply branch protection before first PR), PRs won't merge or BLOCKED, AI reviewer pushback, auto-merge fails for Dependabot/Renovate, branch protection or rulesets, CI fails, authoring reusable workflows, harden-runner, or CODEOWNERS/PR templates.

Guided checklist for adding a new AI provider to gitai's AIService

REST API design patterns including resource naming, status codes, pagination, filtering, error responses, versioning, and rate limiting for production APIs.

Architectural decision-making framework. Requirements analysis, trade-off evaluation, ADR documentation. Use when making architecture decisions or analyzing system design. When making architectural decisions, evaluating trade-offs, writing ADRs, or analyzing system design. NOT for direct code implementation.

Bash/Linux terminal patterns. Critical commands, piping, error handling, scripting. Use when working on macOS or Linux systems. When working on macOS or Linux systems, writing bash scripts, or using terminal commands. NOT for Windows/PowerShell environments.

Apply operations across multiple files simultaneously. Pattern-based bulk modifications, search-and-replace across codebases, consistent changes to many files at once. When the user needs to change multiple files with the same pattern, rename across a codebase, add imports to many files, update versions, or apply consistent modifications. NOT for single-file edits.

AI operational modes (brainstorm, implement, debug, review, teach, ship, orchestrate). Use to adapt behavior based on task type. When adapting AI behavior for specific task types: brainstorm, implement, debug, review, teach, ship, or orchestrate modes.

Decision guide for delegating to caveman-style subagents. Tells the main thread WHEN to spawn `cavecrew-investigator` (locate code), `cavecrew-builder` (1-2 file edit), or `cavecrew-reviewer` (diff review) instead of doing the work inline or using vanilla `Explore`. Subagent output is caveman-compressed so the tool-result injected back into main context is ~60% smaller — main context lasts longer across long sessions. Trigger: "delegate to subagent", "use cavecrew", "spawn investigator/builder/reviewer", "save context", "compressed agent output".

Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.

Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"

Quick-reference card for all caveman modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /caveman-help, "caveman help", "what caveman commands", "how do I use caveman".

Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

Show real token usage and estimated savings for the current session. Reads directly from the Claude Code session log — no AI estimation. Triggers on /caveman-stats. Output is injected by the mode-tracker hook; the model itself does not compute the numbers.

This skill should be used when the user asks about "worktree", "worktrees", "sessões paralelas com worktree", "--worktree flag", "EnterWorktree", "ExitWorktree", ".worktreeinclude", "WorktreeCreate hook", "WorktreeRemove hook", "worktree.baseRef", "isolate subagents", "isolation worktree", "sessões isoladas no Claude Code", "rodar Claude em paralelo com worktrees", or any question about running parallel Claude Code sessions in isolated git worktrees.

Pragmatic coding standards - concise, direct, no over-engineering, no unnecessary comments Always active for ALL code writing. Enforces concise, direct coding standards, testing pyramid, and performance best practices.

Manage and compress conversation context in long sessions. Detect when context is growing large, summarize completed work phases, archive old findings while preserving key decisions. Prevents context degradation. When a session has 20+ turns, when context feels repetitive, when the agent is losing track of earlier work, or when the user says 'summarize what we've done'. NOT for short sessions.

Advanced multi-agent orchestration with parallel workers, synthesis protocols, and coordinator lifecycle. Use when complex tasks require multiple agents working in parallel with intelligent result synthesis. When the user needs multi-agent coordination, parallel task execution, complex multi-domain work, or when /coordinate or /orchestrate is invoked. NOT for single-domain tasks.

Database design principles and decision-making. Schema design, indexing strategy, ORM selection, serverless databases. When designing database schemas, choosing ORMs, planning migrations, or optimizing queries. When working with Prisma, Drizzle, or SQL files.

Production deployment principles and decision-making. Safe deployment workflows, rollback strategies, and verification. Teaches thinking, not scripts. When deploying to production, planning rollback strategies, or setting up CI/CD pipelines. Use with /deploy workflow.

Documentation templates and structure guidelines. README, API docs, code comments, and AI-friendly documentation. When writing README files, API documentation, code comments, or generating AI-friendly documentation.

Patterns for robust error handling across TypeScript, Python, and Go. Covers typed errors, error boundaries, retries, circuit breakers, and user-facing error messages.

Git workflow patterns including branching strategies, commit conventions, merge vs rebase, conflict resolution, and collaborative development best practices for teams of all sizes.

Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools.

Automatic agent selection and intelligent task routing. Analyzes user requests and automatically selects the best specialist agent(s) without requiring explicit user mentions. Always active. Automatically selects the best specialist agent for each user request without explicit user mentions.

Automatic quality control, linting, and static analysis procedures. Use after every code modification to ensure syntax correctness and project standards. Triggers onKeywords: lint, format, check, validate, types, static analysis. When running linters, type checkers, or code formatters. After any code change that needs quality validation.

Lint and format the codebase, then fix the issues found. Use when asked to lint, format, clean up style, or resolve linter/formatter errors.

Persistent cross-session memory management. Enables agents to remember user preferences, project conventions, and past decisions across different sessions using a structured MEMORY.md index and topic files. When the user says 'remember this', 'save this for later', 'don't forget', or when starting a new session and needing to recall past context. Also when /remember workflow is invoked.

Performance profiling principles. Measurement, analysis, and optimization techniques. When diagnosing performance issues, running Lighthouse audits, analyzing bundle size, or optimizing Core Web Vitals.

Whole-repo audit for over-engineering. Like ponytail-review, but scans the entire codebase instead of a diff: a ranked list of what to delete, simplify, or replace with stdlib/native equivalents. Use when the user says "audit this codebase", "audit for over-engineering", "what can I delete from this repo", "find bloat", "ponytail-audit", or "/ponytail-audit". One-shot report, does not apply fixes.

Harvest every `ponytail:` comment in the codebase into a debt ledger, so the deliberate shortcuts and deferrals ponytail leaves behind get tracked instead of rotting into "later means never". Use when the user says "ponytail debt", "/ponytail-debt", "what did ponytail defer", "list the shortcuts", "ponytail ledger", or "what did we mark to do later". One-shot report, changes nothing.

Quick-reference card for all ponytail modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /ponytail-help, "ponytail help", "what ponytail commands", "how do I use ponytail".

Code review focused exclusively on over-engineering. Finds what to delete: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. Use when the user says "review for over-engineering", "what can we delete", "is this over-engineered", "simplify review", or invokes /ponytail-review. Complements correctness-focused review, this one only hunts complexity.