org-universal-validator
Validate skills and plugins with deterministic evidence bundles and strict schema gates. Use when auditing changes or enforcing compliance. Trigger with 'run validation' or 'audit validators'.
When & Why to Use This Skill
The Universal Validator is a robust Claude skill designed to ensure the reliability and compliance of AI skills and plugins through deterministic evidence bundles. It streamlines the auditing process by enforcing strict schema gates and generating comprehensive validation reports, logs, and JSON summaries. This tool is essential for developers and teams who need to maintain high-quality standards and verifiable performance in agentic workflows.
Use Cases
- Compliance Auditing: Automatically generate reviewable evidence bundles to verify that skill or plugin changes meet organizational security and structural standards.
- Automated Quality Assurance: Run deterministic test suites with specific profiles to produce consistent, ground-truth logs and performance reports for any repository.
- Schema Enforcement: Use strict schema gates to validate the structure of new plugins before deployment, ensuring they integrate seamlessly without structural errors.
- Regression Testing: Execute standardized validation procedures during the CI/CD process to detect breaking changes in agent logic or data handling.
| name | org-universal-validator |
|---|---|
| description | "Validate skills and plugins with deterministic evidence bundles and strict schema gates. Use when auditing changes or enforcing compliance. Trigger with 'run validation' or 'audit validators'." |
| allowed-tools | "Read,Write,Glob,Grep,Bash(python:*),Bash(bash:*),Bash(pytest:*)" |
| version | "0.1.0" |
| author | "Your Team <team@example.com>" |
| license | MIT |
Universal Validator
Purpose
Produce deterministic, reviewable validation evidence (reports + JSON + logs) for a repo, plugin, or skill.
Overview
- Phase prompts live in
{baseDir}/agents/ - Procedures live in
{baseDir}/references/ - Deterministic runner + profiles live in
{baseDir}/scripts/
Prerequisites
- Python 3.11+
- Your repo validators and tests installed
Instructions
- Choose a target (repo/plugin/skill).
- Run the deterministic suite with a profile:
python {baseDir}/scripts/run_validator_suite.py \
--target . \
--project example \
--out reports/example \
--profile default
- Optionally run tests:
python {baseDir}/scripts/run_validator_suite.py \
--target . \
--project example \
--out reports/example \
--profile default \
--run-tests
Output
reports/<project>/<timestamp>/summary.jsonreports/<project>/<timestamp>/report.mdreports/<project>/<timestamp>/checks/*.log
Error Handling
- If schema checks fail, fix structure before running tests.
- Treat logs as ground truth; update narratives to match evidence.
Examples
python {baseDir}/scripts/run_validator_suite.py --target . --project pr-1234 --out reports/pr-1234 --profile default
Resources
- Agents:
{baseDir}/agents/ - Procedures:
{baseDir}/references/ - Scripts:
{baseDir}/scripts/