01-cover-letter
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data such as Terms of Reference (ToR), Request for Proposals (RFP), and advertisements. This creates an indirect prompt injection surface where malicious instructions embedded in those documents could attempt to influence the agent's output.
- Ingestion points: External procurement materials (SKILL.md, Required Inputs section).
- Boundary markers: The skill lacks explicit delimiters or warnings for the agent to ignore instructions embedded within these external documents.
- Capability inventory: The skill is restricted to 'read and draft' capabilities and explicitly prohibits high-risk actions like signing or submitting documents (SKILL.md, Capability and Permission Boundaries section).
- Sanitization: No input sanitization or validation of the external documents is performed.
- [SAFE]: The skill includes internal references to other skills within the same repository structure, which are used for routing and quality gates. These are standard navigation patterns and do not involve remote code execution or external downloads.
Audit Metadata