guspatagonico

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. Trigger on "/caveman:compress <filepath>" or "compress memory file".

Quick-reference card for caveman modes and commands. Trigger on "/caveman-help", "caveman help", "what caveman commands", or "how do I use caveman".

Ultra-compressed code review comments. One-line findings with location, problem, fix. Trigger on "review this PR", "code review", "review the diff", or "/caveman-review".

Ultra-compressed communication mode. Trigger on "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or "/caveman".

Help users discover and install agent skills. Trigger on "find a skill", "is there a skill", "how do I do X", or requests to extend capabilities.

You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, decisions, and architectural rules in .brv/context-tree. Uses a configured LLM provider (default: ByteRover, no API key needed) for query and curate operations.

Audit knowledge freshness and coverage. Checks what's documented against the current codebase, identifies stale or outdated knowledge, finds gaps, and provides targeted brv curate commands to fix them.

Execute a phase plan task by task with verification. Loads the plan from the knowledge base, implements each task sequentially, verifies against success criteria, records progress, and transitions to the next phase. Stores execution results via brv curate.

Define project milestones with goals, scoped requirements, and phased roadmaps. Gathers user intent through questioning, assigns REQ-IDs for traceability, derives phases using goal-backward analysis, and optionally captures implementation decisions for gray areas. Stores everything via brv curate.

Interactive knowledge-driven project onboarding. Queries comprehensive knowledge across all domains, presents a structured project overview, identifies knowledge gaps, and suggests running explore to fill them.

Check project progress, resume paused work, and route to the next action. Loads project state from the knowledge base, presents a status report, detects incomplete work or handoffs, and suggests the most logical next workflow to invoke.

You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, decisions, and architectural rules in .brv/context-tree. Uses a configured LLM provider (default: ByteRover, no API key needed) for query and curate operations.

Create structured implementation plans informed by existing knowledge. Queries architecture, conventions, and related implementations, then produces a goal-backward task breakdown with dependencies. Stores the plan via brv curate.

You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, decisions, and architectural rules in .brv/context-tree. Uses a configured LLM provider (default: ByteRover, no API key needed) for query and curate operations.

Automates semantic releases from the previous git tag. Trigger when user asks to bump version, decide patch/minor/major from changes since last tag, create release commit+tag, generate full changelog/release notes, or create GitHub Release with auto-generated notes. Keywords: "subir versión", "bump", "release", "semver", "último tag", "changelog", "release notes", "commit + tag", "github release". Default dry-run. Confirm only for major. Push is manual by default with optional push dry-run/apply modes. Optional `--github` flag creates a GitHub Release with detailed release notes. Examples: "prepará release semver", "decidí semver según cambios", "creá commit y tag de release", "subí versión según cambios desde último tag", "generar changelog completo", "aplicá release con github release".