fix: progressive skill disclosure in claude-code adapter#20
Merged
shreyas-lyzr merged 1 commit intomainfrom Mar 15, 2026
Merged
fix: progressive skill disclosure in claude-code adapter#20shreyas-lyzr merged 1 commit intomainfrom
shreyas-lyzr merged 1 commit intomainfrom
Conversation
Switch claude-code adapter from loadAllSkills() to loadAllSkillMetadata() so CLAUDE.md only inlines skill name, description, and allowed-tools with a file path reference instead of full SKILL.md instructions. Claude Code can read the full instructions on demand, saving ~5000 tokens per skill. Closes #19 Co-Authored-By: Claude Opus 4.6 <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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
claude-codeadapter fromloadAllSkills()toloadAllSkillMetadata(), soCLAUDE.mdonly inlines skill name, description, and allowed-tools with a file path reference — instead of the fullSKILL.mdinstructionsallowedToolsfield toSkillMetadatainterface so metadata-only loading still surfaces tool restrictionssystem-promptadapter is intentionally unchanged — it produces standalone prompts for LLMs without filesystem access, so full inline is correct thereSaves ~5000 tokens per skill in the CLAUDE.md system prompt. Claude Code reads the full
SKILL.mdon demand when the skill is activated.Closes #19
Test plan
npm run buildcompiles without errorsnpm testpassescd examples/standard && gitagent export --format claude-code— outputs skill metadata + file paths, not full instructionscd examples/standard && gitagent export --format system-prompt— still outputs full instructions (unchanged)🤖 Generated with Claude Code