- Home
- Skills
- Mattgierhart
- Prd Driven Context Engineering
- Ghm Gate Check
ghm-gate-check_skill
- Python
17
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill mattgierhart/prd-driven-context-engineering --skill ghm-gate-check- SKILL.md4.1 KB
Overview
This skill validates gate criteria before advancing a PRD lifecycle version. It runs automatically when moving from v0.X to v0.Y or on explicit /ghm-gate-check invocation and returns a concise pass/block summary with any missing artifacts and actionable recommendations. The output is designed to be traceable and machine-readable for downstream automation.
How this skill works
The skill reads the target gate section from the PRD, extracts required criteria and artifact types, then checks the source-of-truth for each artifact ID and status. It verifies cross-references (for example CFD → BR links), assesses handoff readiness for downstream agents, and generates a structured report listing criteria status, missing artifacts, and next steps. It never creates artifacts or override blocks; it only reports findings.
When to use it
- Before advancing a PRD from one minor version to the next (v0.X → v0.Y).
- When a reviewer or CI flow invokes /ghm-gate-check explicitly.
- During milestone sign-off to confirm commercial, technical, and test readiness.
- As a gating step in automated release or documentation pipelines.
- When you need a traceable list of missing artifacts and specific IDs to fix.
Best practices
- Keep PRD gate sections current so extraction yields accurate criteria.
- Use stable artifact IDs (BR-XXX, UJ-XXX, API-XXX, CFD-XXX) and store them in the SoT.
- Ensure artifact statuses are updated from Draft to Active before re-checking.
- Include explicit cross-references in artifacts to enable evidence chain validation.
- Treat a BLOCK as an actionable backlog item list rather than a judgment.
Example use cases
- A product manager runs the check before promoting v0.3 → v0.4 to confirm BR and pricing artifacts exist.
- A CI pipeline triggers the skill to gate advancement and fail the pipeline if key API contracts are missing.
- A handoff to an engineering agent verifies EPIC blockers are cleared and README is synchronized.
- A launch readiness review validates metrics tracking and deployment artifacts before v0.9 → v1.0.
FAQ
No. It only reports missing or incomplete artifacts and provides recommendations for remediation.
How does it determine PASS vs BLOCK?
PASS requires all extracted criteria evaluated with traceable evidence IDs and no open handoff blockers; any missing artifact or unresolved blocker results in BLOCK.