vendor-cyber-risk-analyst
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define workflows for analyzing external, untrusted data provided by third-party vendors, which could contain malicious instructions designed to influence the agent's assessment logic.
- Ingestion points: Ingestion of vendor-supplied security questionnaires (SIG, CAIQ, custom portals) and evidence reports (SOC 2, ISO 27001, penetration test summaries) as detailed in
references/questionnaire_scoring.mdandreferences/evidence_and_attestation_review.md. - Boundary markers: Absent. The skill lacks instructions for using delimiters or explicit 'ignore embedded instructions' warnings when processing external vendor text.
- Capability inventory: None. The skill does not include any tools or scripts for network access, file-system modification, or command execution.
- Sanitization: Absent. There are no guidelines for validating or filtering the content of ingested vendor reports to prevent prompt-based attacks.
- [NO_CODE]: The skill is comprised solely of markdown documentation and guidance. No executable code (.py, .js, .sh), dynamic context injections (!
command), or package manifest files were detected.
Audit Metadata