Strip execution fields from public skillshelf.yaml sidecars#82
Merged
Conversation
The execution layer (inputs, runtime, output, faq) has moved to private configs in the site repo as part of the public/private sidecar split. Public sidecars now contain only catalog metadata: version, category, level, primitive, platforms, tags, author, dates, and certified. Fields removed from all 19 skills: - conversation_starter (moved to SKILL.md conversation flow) - faq (moved to private execution config) - consumes (replaced by skill_output inputs in execution config) - input_schema (replaced by inputs in execution config) - output_schema (replaced by output in execution config) Also updates: - validate_skillshelf_metadata.py: remove consumes/conversation_starter validation - benchmark/loader.py: remove conversation_starter reading from sidecar - CONTRIBUTING.md: rewrite field tables to catalog-only fields - skillshelf-yaml-reference.md: remove Conversation Starter and Consumes sections - skill-authoring-guide.md: replace conversation_starter guidance - conventions-checklist.md: remove faq checkbox Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
timctfl
added a commit
that referenced
this pull request
Apr 8, 2026
The sidecar was accumulating fields that duplicate information already in the SKILL.md body or that are better managed by the website. This strips it to catalog-only metadata: version, category, level, primitive, platforms, tags, author, dates, and certified. Fields removed from all 19 skills: - conversation_starter (defined in the SKILL.md conversation flow) - faq (managed by the website) - consumes (managed by the website) - input_schema (removed, loosely typed and unused) - output_schema (removed, loosely typed and unused) Also updates: - validate_skillshelf_metadata.py: remove consumes/conversation_starter validation - benchmark/loader.py: remove conversation_starter reading from sidecar - CONTRIBUTING.md: rewrite field tables to catalog-only fields - skillshelf-yaml-reference.md: remove Conversation Starter and Consumes sections - skill-authoring-guide.md: replace conversation_starter guidance - conventions-checklist.md: remove faq checkbox Co-authored-by: Tim Petrella <tpetrella@gmail.com> Co-authored-by: Claude Opus 4.6 (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
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.
Summary
conversation_starter,faq,consumes,input_schema, andoutput_schemafrom all 19 publicskillshelf.yamlfilesbenchmark/loader.pyto stop readingconversation_starterfrom the sidecarCONTRIBUTING.mdfield tables and example to reflect catalog-only metadataskillshelf-yaml-reference.md(both write-skill and adapt-skill copies)conversation_starterguidance inskill-authoring-guide.mdfaqcheckbox fromconventions-checklist.mdWhy
The sidecar was accumulating fields that duplicate information already in the SKILL.md body or that are better managed by the website. Stripping it to catalog-only metadata keeps the contributor surface simple: category, level, platforms, tags, author, version.
Test plan
python scripts/validate_skillshelf_metadata.py skills/*/passes (19 skills, exit 0)🤖 Generated with Claude Code