Categories
Indexed Skills (7)
chain-metrics
Display chain metrics summary after a completed agent chain. Use when the user asks for cost/token/duration breakdown of the last chain run.
commit
Create a conventional commit. Analyzes staged/unstaged changes, drafts a commit message, and commits. Use when the user says "commit", "commitni", or asks to save changes.
deep-analysis
Deep analysis of project structure, logic, patterns, and dependencies. Use when the user wants a thorough understanding of how the project works, its architecture, or specific subsystems.
push
Push current branch to remote. Checks branch state and remote tracking before pushing. Use when the user says "push", "pushni", or asks to push changes.
re-review
Re-run the review chain on existing code without implementing anything. Use when the user wants a fresh security or architecture review of code that is already written.
tier-check
Analyze a task and recommend the appropriate tier (0-4) with the full agent chain. Use when the user wants to know which tier a task falls into before starting work.
bootstrap
Run the bootstrap protocol to customize all [PROJECT-SPECIFIC] sections for a new project. Use when setting up the framework for the first time or when the user says "bootstrap".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.