spec-verification-criteria

To verify agent specifications, you check instruction files against 7 critical verification criteria and independence rules. This process validates formatting, content, and rule-based requirements to ensure AI governance standards are met before deployment.

The 7 verification criteria are a defined set of compliance checks used to validate agent instruction records and associated metadata. They ensure specifications meet formatting, content, and independence rules to prevent deployment errors.

Yes, spec verification uses TypeDB to retrieve and compare agent artifacts like instruction records and specifications. This ensures accurate validation against DLMS standards and the 7 predefined verification criteria.

To prevent an AI agent from verifying its own work, the verification process enforces independence rules. This assertion ensures system integrity by confirming that the verifying agent is distinct from the agent being checked.

Generating a verification record requires access to TypeDB and specific agent artifacts like instruction files and specifications. The Skill retrieves these artifacts, compares them against 7 criteria, and outputs a detailed pass/fail report.

You should run specification validation before deploying a new AI agent. This compliance check flags discrepancies in instruction files and associated metadata, ensuring all DLMS standards and independence rules are satisfied to maintain system integrity.