Skip to content

Add charles-claude-skills to Agent Skills#273

Open
cabaynes wants to merge 1 commit into
jqueryscript:mainfrom
cabaynes:add-charles-claude-skills
Open

Add charles-claude-skills to Agent Skills#273
cabaynes wants to merge 1 commit into
jqueryscript:mainfrom
cabaynes:add-charles-claude-skills

Conversation

@cabaynes
Copy link
Copy Markdown

What

Adds charles-claude-skills to the 🧠 Agent Skills section.

What's in the skill bundle

  • Paired /checkpoint + /resume — session-continuity across Claude Code context resets. /checkpoint writes a structured handoff file; /resume reads it in a fresh session. Shipped as a paired bundle (they share a folder, must install together).
  • /newproject — parameterized workspace project bootstrapping. Reads $WORKSPACE_DIR (default ~/projects/), idempotent, optional git init + umbrella-file integration.
  • /skill-dict — manage a personal catalog of installed Claude Code skills. Five subcommands: list, show <name>, sync, add <name>, check-updates.

Why I'd like it listed

These were evaluated against current Anthropic best practices before release:

  • Static review against an 18-rule rubric from writing-skills + skill-creator.
  • Trigger-accuracy benchmark following skill-creator's description-optimization methodology: 20 queries per skill (10 should-trigger + 10 adversarial should-not-trigger), LLM-as-judge scoring.
  • Results: 100% precision + 100% recall across all four skills. Full report in eval-results.md.

Repo includes top-level + per-skill READMEs, MIT LICENSE (root + per-skill), INSTALL.md with copy-vs-symlink guidance, CONTRIBUTING.md with the eval methodology so others can verify changes don't regress.

Entry placement

Added at the bottom of the no-emoji block under Agent Skills (zero stars, fresh release). Happy to relocate if the maintainer prefers a different position.

A paired /checkpoint + /resume bundle for session-continuity across context
resets, plus standalone /newproject and /skill-dict utilities. Evaluated
against Anthropic's writing-skills + skill-creator rubric to 100% trigger
accuracy on a 20-query precision/recall benchmark per skill.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant