project-architecture-quality-loop
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and processes external data from the project's source code and generated assessments. 1. Ingestion points:
docs/architecture/ASSESSMENT.mdand module digests from thecode-analyzer. 2. Boundary markers: No explicit delimiters or instructions are provided to the model to ignore potential instructions embedded in the analyzed code. 3. Capability inventory: The skill has the ability to write to the file system and dispatch sub-agents that can modify code. 4. Sanitization: There is no automated sanitization of ingested content, but the skill includes mandatory human-in-the-loop approval gates for consequential changes. - [SAFE]: The skill includes several security best practices. It is configured with
disable-model-invocation: trueto ensure it only runs when explicitly called by the user. Furthermore, it strictly gates substantial and structural changes behind mandatory user approval and uses automated test verification to maintain the integrity of the project's baseline.
Audit Metadata