production-diagrams
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external project artifacts which serves as an entry point for untrusted data.
- Ingestion points: Step 1 involves reading multiple project files including 'standalone contract', 'core contract', and 'canon' (SKILL.md).
- Boundary markers: The instructions lack specific delimiters or guardrails to ignore instructions embedded within the source artifacts.
- Capability inventory: The agent is instructed to write JSON and Markdown files to the 03_preproduction/diagrams/ path (SKILL.md).
- Sanitization: There are no sanitization steps defined for the content extracted from external artifacts.
Audit Metadata