ai-assessment-scale
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external repository content, README files, and documentation via the
project_url_or_codebaseinput, establishing an indirect prompt injection surface. - Ingestion points: External codebase content, commit histories, and documentation are analyzed during the 'Project Discovery' and 'Evidence Analysis' phases described in
SKILL.md. - Boundary markers: The skill includes explicit defensive instructions for the agent to use 'Delimiter isolation,' treating external inputs as
<untrusted-content>and ensuring skill instructions take precedence over data. - Capability inventory: The skill primarily uses data-reading capabilities to generate structured assessment reports; no high-risk capabilities like network exfiltration or shell execution are requested for processing this data.
- Sanitization: A 'Security Notice' section provides specific guidance for the agent to detect and disregard injection patterns such as 'ignore previous instructions' or 'you are now' when found within project content.
Audit Metadata