feat: add skill-writer reference skill#3
Open
vtri950 wants to merge 2 commits into
Open
Conversation
📋 Collection Validation Results✅ All 5 collection(s) validated successfully! Python unit tests✅ all passing — 134/134 passed across 9 file(s). |
Fill the two gaps in the skill-writer reference: document the physical anatomy (folder naming, frontmatter fields, disclosed files) and add a miniature bad->good worked example applying no-op cut, leading word, and disclosure in one pass. Keep it agent-agnostic — only name/description/ disable-model-invocation are documented, extra fields deferred to each agent's docs. Align README wording to match. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Adds a new skill-writer reference skill — a style guide for writing agent instructions (skills) that the agent follows the same way every run, without wasting tokens. Model-agnostic: works with GitHub Copilot and any other agent.
Type of Change
Related Issues
Changes Made
skills/skill-writer/SKILL.md— the principles reference for writing predictable, token-efficient skills (invocation choices, description writing, scope gates, information hierarchy, leading words, pruning)skills/skill-writer/GLOSSARY.md— the full vocabulary of core terms, surfaced via progressive disclosure and linked fromSKILL.mdskills/skill-writer/README.md— the human-facing intro: the standing-instruction-card framing, the one big idea (same way, not same output), a section-by-section walkthrough, and a problem→teaches tablecollections/skill-writer.collection.yml— collection manifest registering the skill and wiring the README viareadme.path, matching the existing ospo / skill-eval / foursight collectionsTesting
Test Coverage
npm run validate)npm run skill:validate)Manual Testing Steps
npm run skill:validate→✅ skill-writer is valid(1/1 skills)npm run validate→✅ All 5 collection(s) validGLOSSARY.mdlink fromSKILL.mdand the README link from the collection manifest resolveChecklist
Additional Notes
The skill lives at the top-level
skills/directory, which the README documents as the skills location and whichvalidate-skillsscans.package-lock.jsonwas intentionally excluded from this PR.By submitting this pull request, I confirm that my contribution is made under the terms of the project's license.