Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Generate before/after PNG screenshots for a YAS branch's rendering changes, push them to the yas-pr-screenshots repo, and hand back a markdown before/after table for the PR description. Use when the user wants real image screenshots (not ANSI-stripped text) attached to a YAS pull request, or asks to shoot/capture before-after statusline screenshots for a branch.
How to read files (audio, documents, images) that Sir sent to Alfred on any channel — Slack file uploads, Telegram voice memos, MMS audio, generic URLs. Converts file references into transcripts + vault records so Alfred doesn't say "I don't see your file".
Add or regenerate 16×16 item/resource icons on items-sheet.png (Pillow), align with game-shared asset coords, InteractableTexts, and optional interactableDisplayName. Use when adding new pickup sprites, patching the items atlas, or fixing misaligned tiles.
Finalize documentation and project metadata for a ship-ready release. Use after implementation is complete, tests pass, and devcheck is clean. Safe to run at any stage — each step checks current state and only acts on what still needs work.
Multi-shot narrative & character consistency — a character bible with a locked front-portrait anchor, view-matched reference selection, recent-frame carry-forward, Cameo (a user photo as the lead), and global planning for long scripts/novels. Trigger on top of the generation line when the SAME character must look the same across many shots/scenes, or for long-form story→video; do NOT trigger for one-off single-shot generation.
The decision layer for real footage — understand → select → produce an EVIDENCE-bearing rough cut. Trigger when the EDIT task is "find / select / reduce / clean" (remove dead air, drop fillers, pick highlights, cut 1 hour to 3 minutes), NOT executing a known timecode edit (that is stage-edit). Deterministic auto-cuts (silence/filler/quality) are reliable; narrative/emotional selection is a low-confidence DRAFT for the user to review.
Deterministic editing of real user-supplied footage — probe inputs, then trim / concat / burn-subtitles / overlay / mix by explicit timecodes via an edit_decisions or plan.json timeline, with `ovs edit`. Trigger for clip-factory / montage / repurpose / localization / adding narration to existing footage; do NOT trigger for designed HTML (stage-compose) or AI-generated footage (stage-generate).
Process raw inbound content (emails, voice memos, notes) into structured Obsidian vault records with proper frontmatter, wikilinks, and file placement.
Read operational vault records and extract latent knowledge into structured learning records with proper frontmatter, wikilinks, and file placement.
Fix vault quality issues — broken frontmatter, invalid values, orphaned records, garbage content.