graphite-stack
Split changes into a Graphite stack with PR titles and descriptions. Use when asked to create or propose a stacked PR series.
Browse published AI skills maintained by Metta-AI. Compare skill descriptions and follow each source repository for setup instructions.
Indexed skills
45Repository stars across indexed skills
90Repository forks
45Average repository stars per indexed skill
2Repository stars may repeat across skills from the same repository.
Split changes into a Graphite stack with PR titles and descriptions. Use when asked to create or propose a stacked PR series.
Sync the branch with origin/main by fetching and merging or rebasing, then resolve conflicts and summarize changes. Use when asked to sync with main.
Resolve merge conflicts with repo-specific priorities.
Quick repo state and history scan.
Review branch changes vs main using a merge-base diff and return prioritized findings, risks, and cleanup opportunities. Use when asked to review changes against main/origin/main.
Generate a PR title and description from the branch diff vs main. Use when asked for a PR summary or PR description.
Diagnose pytest or CI failures, identify root cause, and implement the minimal fix. Use when tests fail or CI reports errors.
Audit scripted roles and behavior coverage.
Run the standard local validation steps for tribal-village (alias: tv-validate).
Investigate tint layers and frozen tiles.
Read Codex logs for this repo and synthesize docs in docs/.
Explain maps, variants, and curriculum ordering used by a training command or recipe. Use when asked to audit a recipe configuration.
Simplify code and remove legacy/backcompat shims while preserving behavior. Use when asked to clean up or simplify code.
Make an existing change more concise without altering behavior. Use when asked to simplify or reduce a diff.
Compare Nim and Python scripted agent implementations and align behavior. Use when asked to port or ensure parity between Nim and Python.
Create a plan to modularize the AI include chain.
Locate the latest checkpoint for a run id and provide a follow-up play/eval command. Use when asked to find a checkpoint.
Generate a list of new behaviors aligned to territory control.
Investigate cliffs, ramps, and elevation traversal.
Audit for cleanup or complexity hotspots.
Run lint autofix (default `metta lint --fix`), address remaining lint errors, and summarize changes. Use when asked to fix lint issues.
The default order uses GitHub repository stars from our indexed data. Stars belong to the source repository and may be shared by multiple skills. They indicate repository interest, not skill installs or an independent quality score.