measure-survey-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of raw survey response rows and open-text excerpts, which could contain malicious instructions intended to bypass the agent's logic or influence its analysis findings.
- Ingestion points: The skill accepts raw survey data and open-text responses as primary inputs (documented in
SKILL.md). - Boundary markers: Absent. The instructions do not mandate the use of delimiters or explicit 'ignore instructions' markers when processing the input data.
- Capability inventory: The skill is intended to use
ReadandGreptools to generate markdown reports; while no network or write tools are specified in the instructions, the lack of platform-level restrictions (frontmatter) implies the agent's full toolset might be available if not properly scoped. - Sanitization: Absent. No sanitization or validation of the survey content is requested before it is clustered or quoted in the final report.
Audit Metadata