Pmo governance

Skill witchwasin/Axiom-PMO/skills/pmo-governance

Use for RAID, decisions, traceability, risk, policy checks, and Strict-mode guardrails.From its SKILL.md

Install

npx -y skills add witchwasin/Axiom-PMO --skill pmo-governance

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.7 KB, as published. Nobody here has run it

pmo-governance

Purpose

Maintain governance consistency without turning Lite projects into heavyweight programs.

Trigger

Use for risk triage, decision logging, RTM checks, approval policy, blocker handling, Strict-mode escalation, Change Control, or Externalization review.

Required Inputs

PROJECT.md, relevant governance artifacts, and runtime config from pmo-config/context-map.json, pmo-config/policy.json, pmo-config/artifact-policy.json, pmo-config/reference-types.json, and pmo-config/validation-rules.json.

Allowed Context

Read only governance artifacts needed for the gate. Do not load archived skills.

Mode Behavior

Use pmo-config/artifact-policy.json for required/conditional/optional governance artifacts and pmo-config/policy.json for strict triggers, approval roles, and evidence values. Lite logs only material decisions/risks; Standard logs meaningful risk or business decision; Strict follows the configured governance artifact contract.

Execution Steps

Classify risk and strict triggers.
Check approval gates and evidence.
Verify RTM coverage for Strict.
Confirm config and docs do not drift.
Validate CHANGE-REQUESTS.json when a real deviation exists; AI-proposed impact is candidate evidence, not exhaustive discovery or approval.
For Externalization, maintain EXTERNALIZATION.json as a registry of what leaves the governed boundary: exact outgoing artifact paths and digests, declared classification, minimization/redaction, deterministic scan result, and truthful network-transfer declaration. Confidential/Restricted transfers and non-clean scans require a named Human reviewer and a resolvable DEC-###; never declare such content safe on your own, never echo a detected value in a diagnostic.
For Research, keep RESEARCH/RESEARCH.md and RESEARCH/PROVENANCE.json source-backed and draft Impact Assessment and Change Proposals as candidate evidence. Only a Human accepts proposals at Scope (RESEARCH-004); an unresolved accepted-impact proposal blocks Scope (RESEARCH-005).

Output Contract

Return risk status, required artifacts, missing evidence, and policy deviations.

Approval Rules

Human approval is required for scope, design where applicable, release, mode escalation, and production-sensitive decisions.

Validation Command

node cli/axiom.mjs doctor

Prohibited Actions

Do not downgrade Strict triggers silently, fabricate decision evidence, hardcode artifact matrices, log every minor AI action, or approve on behalf of a human.

Completion Criteria

Governance artifacts match mode requirements and doctor/validator checks pass.