nix-config

Use when writing technical articles, blog posts, X/Twitter threads, LinkedIn posts, documentation, runbooks, or any text that must sound like a specific person, not a generic assistant. Also use when the user asks to humanize text, remove AI-sounding language, rewrite in their voice, or convert rough notes into publishable content. Trigger even when the user just says "write this up" or "draft a post about X" without explicitly mentioning voice or tone.

Use when reviewing architecture, planning refactors, reducing coupling, improving testability, or deciding whether modules should become deeper behind simpler interfaces.

Use when the user asks for caveman mode, terse mode, fewer tokens, less prose, or very brief technical communication.

Use for hard bugs, failing checks, pipeline failures, homelab incidents, performance regressions, or reports that something is broken and needs root-cause analysis.

Use before git mutation, branch deletion, resets, cleaning, force pushes, broad restore commands, PR operations, or any git action with destructive or remote side effects.

Use when a plan, design, or implementation request has ambiguity that could change scope, architecture, acceptance criteria, safety, or validation.

Use when work should continue in another session, another lane, or a fresh context window, especially after planning, debugging, or partial implementation.

Use when implementing a feature or bug fix where behavior can be verified through tests, especially when the user mentions TDD, regression tests, red-green-refactor, or integration tests.

Use when creating, revising, or evaluating an agent skill, including skill descriptions, trigger behavior, progressive disclosure, and bundled resources.

Use when the code area is unfamiliar, the user asks for a map, or implementation needs broader module, caller, data-flow, or ownership context before editing.

Use when important outputs need break-first review. Run an inline attacker pass before readiness claims.

Use when decisions span multiple artifacts and drift risk is high. Keep source-of-truth and downstream outputs synchronized.

Use after substantial implementation and before merge to request structured review with explicit diff scope, severity levels, and readiness verdict.

Use when a request is feature design heavy, ambiguous, or likely to change architecture. Turn ideas into an approved design before implementation handoff.

Use when one-pass iteration is likely to miss stronger solutions. Generate bounded variants, score them, and converge with explicit evidence.

Use when handling review feedback so responses are technical, verified, and implementation actions are sequenced safely.

Use when setting up isolated implementation streams for risky or parallel work. Create predictable worktree locations and verify clean baselines.

Use before any completion claim, commit, or PR update. Require fresh command evidence before saying work is done or passing.