Loading...
Loading...
Found 14 Skills
Shepherd a GitHub pull request through AI review bots and CI until it is green on the latest commit: watch or poll the PR, verify every bot finding against the real code, fix what is real, answer each inline bot comment with a VALID / INVALID / OUT OF SCOPE verdict reply (with the fixing commit), react and resolve the thread, answer summary-only nitpicks in one top-level comment, fix CI failures, rerun infrastructure flakes and rebase when needed. Provider-agnostic: works with any review bot (Pullfrog, CodeRabbit, Copilot, Gemini and others). Use whenever the user asks to monitor, watch, babysit, shepherd or land a PR, to get a PR green, or to handle, triage or reply to review-bot comments, even if they don't name this skill.
For the editor's maintainers. Turns what users corrected into lasting fixes in this repo, for every user, enforced by code and tests rather than notes. Reads the suggestions users marked "would help everyone" (~/.ai-video-editor/suggestions.jsonl), taste rules said twice (taste.py report), preview corrections (edits/*/corrections.jsonl), review-page notes (*/review.json) and open GitHub issues labelled feedback; groups repeats; for each item the maintainer accepts, writes the general rule into PRINCIPLES.md and the skill that owns it, adds or tightens the check that enforces it, adds a test that fails before and passes after, runs the full test set and leaves a CHANGELOG.md entry to commit. Use when the maintainer says "improve the editor", "turn feedback into fixes", "process suggestions", "what are users complaining about", "go through the feedback issues". Not for one user's own preferences (that is taste), and not for editing a video.