Validate AI-executed code changes against proposals and architectural constraints.
npx skills add https://github.com/insentek/VAPA --skill vapa-audit
Please help me install this Agent Skill. Skill: vapa-audit Source: https://github.com/insentek/VAPA/tree/main/skills/vapa-audit Command: npx skills add https://github.com/insentek/VAPA --skill vapa-audit
This skill solves the uncertainty of whether an AI-generated implementation is truly ready for production by providing an independent, evidence-based validation gate between execution and human review.
Ask the agent to run the vapa-audit skill to validate the current execution workspace and determine if the changes are ready for a pull request.
High-intent search queries and answers about installing and using this skill.
You can validate AI-generated code for PR readiness by running an audit skill that checks execution workspaces against approved proposals and test reports. It verifies implementation integrity and outputs a formal READY_FOR_PR or NEEDS_REVISION verdict.
Traceability verification for agent-executed code ensures that AI implementations strictly adhere to approved proposals, acceptance criteria, and design constraints. It acts as an evidence-based validation gate before human review.
To generate an automated audit report for AI code changes, the validation process analyzes state files, execution logs, and evidence directories. It outputs a structured VAPA_AUDIT.md report detailing findings and test results.
Yes, you need access to state files, execution logs, and evidence directories to validate your project-local execution workspace. These inputs are required to verify implementation integrity and produce a formal audit verdict.
The best way to check if an AI implementation is ready for production is to use an independent validation gate that compares code changes against architectural constraints. It provides clear status indicators like READY_FOR_PR or NEEDS_REVISION to guide next steps.
Your AI code audit returns a NEEDS_REVISION status when the implementation fails to strictly adhere to approved proposals, test reports, or architectural constraints. It indicates the changes require corrections before moving to human review.
D0NMEGAScan local GSD workspace directories and summarize WORKSPACE.md manifests.
mcevoyinitAnalyze code diffs and produce evidence-based review findings with a verdict.
vrcmsExecute build, type, lint, test, and security checks in a verification pipeline.
insightwavesmithAnalyze code changes to identify those meeting evidence-shape-check criteria.
