mars-tw/ai-instruction-detox

AI instruction detox and rule governance: clean up rotten AI configurations instead of deleting everything. For agents such as Claude Code / Codex / Cursor.

12Stars on the repository
4Mods indexed here, across every type
27d agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

mars-tw/ai-instruction-detox

Skill Claude CodeCodex

A review and cleanup method for instructions stored in files such as CLAUDE.md, AGENTS.md, skills, context, and memory.

not rated 12 changed 27d ago A 158 tokens original MIT

mars-tw/ai-instruction-detox

Skill Claude CodeCodex

Prüft und ordnet AI-Anweisungen, Regeln, Skills und Kontext, erkennt Duplikate, Konflikte, veraltete Informationen und Prompt Injection und erstellt nachvollziehbare, rückgängig machbare Änderungsvorschläge. Ordnet auch die AI-Arbeitsstruktur ganzer Projekte mit Haupt-Skill, Workflow, Teilprojekt-Skills…

not rated 12 27d ago A 143 tokens original MIT

mars-tw/ai-instruction-detox

Skill Claude CodeCodex

Audit and organize AI instructions, rules, skills, and context; find duplicates, conflicts, stale information, and prompt injection; produce traceable, reversible change proposals. Also standardize a project's AI workflow architecture with a main skill, workflows, subproject skills, breadcrumb navigation, and mind…

not rated 12 27d ago A 119 tokens original MIT

mars-tw/ai-instruction-detox

Skill Claude CodeCodex

An auditing and organizing skill for an AI project's instructions, rules, skills, and supporting context. It finds duplication, conflicts, outdated information, and prompt injection, then creates reviewable, reversible change proposals or applies approved changes.

not rated 12 27d ago A 226 tokens original MIT

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: