AI prompts.
Agent skills.

Use when defining governance metrics, reading governance health snapshots, generating the governance dashboard, or understanding the governance data model. Referenced by the governance-auditor agent and the /governance-health command.

Use when conducting a governance audit — detecting semantic drift in governance constraints, inventorying governance debt, checking three-frame alignment, or when the governance-auditor agent needs methodology for deep investigation.

Use when acting as the adversarial spec reviewer — raises steel-manned objections across six categories before plan approval, requires evidence per objection, and discloses what was not challenged

Use when writing governance constraints for HARNESS.md, translating governance language into operational meaning, reviewing existing governance constraints for falsifiability, or when "/governance-constrain" needs guidance on the authoring workflow.

Use when acting as the decision-archaeology agent — surfaces decisions a spec has made (including the silent ones), emits each material choice as a Henney-style pattern story for human disposition, and pays down intent debt before plan approval

Use when the user wants to create or update a Team Topologies Team API document with AI literacy portfolio assessment data — generates a template Team API with literacy levels, discipline scores, shared gaps, and improvement plans, or updates an existing Team API with the latest assessment data

Use when syncing HARNESS.md conventions to Cursor, Copilot, and Windsurf convention files — reads Context and Constraints sections and generates tool-specific output so all AI coding tools share the same project rules

This skill should be used when the user asks about "writing conventions", "codebase context", "HARNESS.md context section", "convention documentation", "how to write enforceable rules", or needs guidance on the Context section of HARNESS.md.

Use when creating new source files, writing new functions or types, or significantly rewriting existing code — ensures code is structured for humans to read first, with narrative preambles, reasoning-based documentation, and presentation ordered by understanding rather than compiler convention

Use when reviewing or refactoring code and wanting a structured lens beyond SOLID — applies Daniel Terhorst-North's CUPID properties to surface improvement opportunities in any codebase or language.

Use when generating a human-readable onboarding document from HARNESS.md, AGENTS.md, and REFLECTION_LOG.md — produces a friendly guide for new team members joining a harnessed project

Use when setting up a new project's conventions, onboarding AI to an existing codebase, after team composition changes, or when AI output quality varies depending on who prompts — guides structured discovery of tacit team knowledge into explicit, enforceable artefacts