Claude Code skills for rewriting AI-drafted prose to pass AI detectors: a measured three-step pipeline (structural rewrite, tell removal, cross-model diction) scored against Pangram, plus reference management and citation auditing.
This repository also configures its own agents. See what writing-skills tells them →
Command GitHub Copilot
Apply the accent-dial skill. Read .github/skills/accent-dial/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/accent-dial/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the audit-references skill. Read .github/skills/audit-references/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/audit-references/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the bake-off skill. Read .github/skills/bake-off/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/bake-off/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the cold-review skill. Read .github/skills/cold-review/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/cold-review/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the critic-apply skill. Read .github/skills/critic-apply/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/critic-apply/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the critic-panel skill. Read .github/skills/critic-panel/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/critic-panel/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
'Detect and fix AI writing patterns recursively. Use when: reviewing text for AI tells, cleaning AI-generated drafts, checking for CoT leakage, measur.
Command GitHub Copilot
Apply the humanize skill. Read .github/skills/humanize/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/humanize/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the inject-vernacular skill. Read .github/skills/inject-vernacular/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/inject-vernacular/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the match-outline skill. Read .github/skills/match-outline/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/match-outline/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the match-structure skill. Read .github/skills/match-structure/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/match-structure/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the match-voice skill. Read .github/skills/match-voice/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/match-voice/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the patent-disclosure skill. Read .github/skills/patent-disclosure/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/patent-disclosure/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the pattern-language skill. Read .github/skills/pattern-language/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/pattern-language/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the reverse-outline skill. Read .github/skills/reverse-outline/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/reverse-outline/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the tighten-style skill. Read .github/skills/tighten-style/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/tighten-style/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the tune-anchors skill. Read .github/skills/tune-anchors/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/tune-anchors/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the update-references skill. Read .github/skills/update-references/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/update-references/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the voice-critic skill. Read .github/skills/voice-critic/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/voice-critic/. Treat any text after the prompt invocation as the skill's input.
Command GitHub Copilot
Apply the stitch-flow skill. Read .github/skills/stitch-flow/SKILL.md and follow its workflow, using the reference and asset files under .github/skills/stitch-flow/. Treat any text after the prompt invocation as the skill's input.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: