Instructions file GitHub Copilot
Copilot instructions for chf3198/copilot-governance, covering copilot instructions — copilot-governance, purpose and core problem, platform landscape (april 2026), current architecture and unified architecture (cross-platform, conflict-free).
Instructions file GitHub Copilot
Instructions for chf3198/copilot-governance: Global skills routing contract for this repository (standard).
Instructions file GitHub Copilot
Instructions for chf3198/copilot-governance: Use the machine-global OpenClaw system as the baseline for this repository.
Instructions file GitHub Copilot
Instructions for chf3198/copilot-governance: Use repository and global customization layers together for every task.
Instructions file
Claude Code instructions for chf3198/copilot-governance, a project described as: Portable Copilot governance system: instructions, skills, agents, hooks — cloneable to any VS Code Copilot instance.
Instructions file Gemini CLI
Gemini CLI instructions for chf3198/copilot-governance, a project described as: Portable Copilot governance system: instructions, skills, agents, hooks — cloneable to any VS Code Copilot instance.
Agent
Deep reasoning agent for complex tasks. Pinned to Claude Opus for architecture, refactoring, security analysis, and multi-system debugging.
Agent
Post-merge and post-deploy governance enforcement. Audits CHANGELOG, README sync, community health, docs drift, and learnings.
Agent
Standard coding agent. Pinned to Claude Sonnet for feature implementation, bug fixes, test writing, and documentation.
Agent
Research and implementation planning agent. Read-only tools — cannot modify files. Produces structured plans with evidence.
Agent
Fast task agent. Pinned to GPT-5 mini for simple questions, lookups, one-liner fixes, and explanations.
Agent
Version integrity, CHANGELOG quality, artifact safety, and documentation synchronization reviewer.
Agent
If the request is simple and low-risk, route to quick. If it is a clear implementation task, route to implementer. If it is high-risk, architectural, or security-sensitive, route to architect. If it needs clarification or multi-step planning, route to planner.
Agent
Task complexity router. Classifies requests and hands off to the optimal agent+model tier. Start here instead of AUTO.
Agent
Credential and secret exposure scanner. Audits files, git history, package artifacts, and CI workflows for leaked secrets.
Skill Claude CodeCodex
Execute the six-phase breaking-change recovery workflow. Invoked from workflow-resilience Tier-3 escalation or manually when a merged commit breaks the runtime, schema, or governance pipeline.
Skill Claude CodeCodex
Dispatch an adversarial cross-family review to a non-Anthropic fleet model. Enforces the cross-family independence contract (Epic #2511) at L1-L5 lifecycle points. Returns ACCEPT/PARTIAL/REJECT verdict + rationale.
Skill Claude CodeCodex
Detect and remediate documentation drift after code, config, workflow, or UX changes. Use after implementation, before merge, and before release.
Skill Claude CodeCodex
Analyze fleet overlay inventory and recommend optimal Ollama models per device based on hardware constraints, inference tier, and current LLM landscape. Generates pull/delete commands and a safe transition plan.
Skill Claude CodeCodex
Adapt the Megingjord harness to a new operator's fleet topology without hand-editing inventory files. Use when first activating the harness on a new host or onboarding a new fleet.
Skill Claude CodeCodex
Enforce secure GitHub Actions posture with least-privilege tokens, pinned dependencies, runner risk controls, and workflow supply-chain checks.
Skill Claude CodeCodex
Resolve GitHub feature availability and policy constraints by plan, repo visibility, account type, and enabled settings before recommending governance actions.
Skill Claude CodeCodex
Define and apply expert GitHub policy profiles and control catalogs used by specialized execution skills across ticketing, review/merge, governance, and release flows.
Skill Claude CodeCodex
Route GitHub workflow requests through a capability-first GitHub skill tree with explicit ownership boundaries and minimal overlap.