third-party-risk-summaries
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is strictly instructional and does not contain any executable scripts, binaries, or configuration files that trigger code execution.
- [PROMPT_INJECTION]: The skill is designed to process external, untrusted documents such as vendor due diligence materials and SOC reports, which establishes an attack surface for indirect prompt injection. 1. Ingestion points: Vendor profiles and due diligence materials are ingested via the agent's context as specified in SKILL.md. 2. Boundary markers: There are no explicit delimiters or instructions provided to separate the system instructions from the untrusted vendor data. 3. Capability inventory: The skill lacks any technical capabilities such as file system writes, network requests, or subprocess execution that could be subverted by a malicious payload in the ingested data. 4. Sanitization: The instructions do not include any steps for sanitizing or validating the contents of the processed materials.
Audit Metadata