Loading...
Loading...
Found 30 Skills
Measure whether a repository's AI instructions still earn their place, by running the same real task many times with the layer intact and with it stripped, then grading every rule against what actually changed. Runs both arms itself in throwaway git worktrees and never touches the working tree. Agent-agnostic across CLAUDE.md, AGENTS.md, .claude/, .agents/, .cursor/rules, .clinerules, .windsurfrules and copilot-instructions. Use when the user wants to prune, audit, clean up, shrink or "delete" their CLAUDE.md, AGENTS.md, cursor rules, agent instructions or AI layer; when they ask whether their rules are still needed, whether their context is bloated, or what to cut; or when they mention ablating, ablation, or testing their agent without its instructions.
Use when lasting learnings from the current session should be recorded in agent instruction files - AGENTS.md, CLAUDE.md, or the user's global instructions - such as pitfalls hit, mistakes corrected, conventions found, or commands that finally worked, whatever the conversation language. Proposes minimal, deduplicated edits per file and applies only the approved ones. Not for general memory, a full audit of instruction files, or improvement requests for skills.
AI/LLM: Use when crafting system prompts, optimizing LLM outputs, or improving agent instructions. NOT for general coding.
Create, audit, shorten, or scope AGENTS.md and CLAUDE.md instructions. Use writing-skills for on-demand skills; excludes human-facing documentation and READMEs.
Improve (or bootstrap) an AGENTS.md / CLAUDE.md file using `<important if>` conditional blocks so the agent actually attends to the right guidance at the right time. Use this skill whenever the user mentions AGENTS.md, CLAUDE.md, agent instructions, project rules for AI, "my claude config", onboarding docs for agents, or asks to tighten / shorten / audit / rewrite an existing one — even if they don't explicitly say the filename. Also use when the user complains that an agent keeps ignoring their project rules.
Use when writing or editing instructions a model will later execute — anything future context windows load as guidance. Triggers on 'edit this skill', 'update CLAUDE.md / AGENTS.md', 'write agent rules', 'improve this prompt'.
Bootstrap/refresh MSG Architect methodology in AGENTS.md; show a review-first diff, validate, then self-remove.
Refine human- or agent-facing prose by removing AI tells, filler, vague abstraction, and instruction noise without changing facts, authority, structure, or intended voice. Use for final cleanup or explicitly requested pruning of verbosity, repetition, and instruction load in technical writing, skills, agent instructions, handoffs, tickets, and reports. Exclude content design, factual review, translation, and code style.
Use a custom te-agent Agent's instructions and dependency assets for a task in a local client such as Codex or Claude Code, using the client's model and local files.
Research, audit, create, update, or reorganize repository AGENTS.md instructions and task-routed supporting guidelines. Use when an agent must ground repository guidance in inspected evidence, keep every new or rewritten AGENTS.md within 100 lines, preserve existing instructions, and obtain approval before writing or destructive changes.
Use when creating a skill or memory, when one fails in practice, or when the user reports one misbehaving or stale.
Write, rewrite, or refresh an AGENTS.md (repo root or one package/module) from the actual codebase — Always / Ask First / Never / Validation Commands rules, task checklists, real paths and commands. Follows a repo template when present, else a built-in default. Use for "write AGENTS.md for this package", "rewrite our AGENTS.md", "refresh AGENTS.md".