interview
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a repository-grounded workflow, prioritizing existing project documentation such as .agents/memory/ and CONTEXT.md to inform the discovery process.
- [SAFE]: There are no indicators of prompt injection, obfuscation, or persistence mechanisms. The skill's instructions focus on structured discovery and planning.
- [SAFE]: The skill limits its attack surface by setting
disable-model-invocation: true, which restricts the agent's ability to call external tools autonomously. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it processes external documentation, links, and transcripts provided by the user (Ingestion points: SKILL.md workflow step 1). While it lacks explicit boundary markers or sanitization for this content, its capabilities are restricted to generating Markdown artifacts in a designated local directory (Capability inventory: references/questionnaire.md), which does not present a significant security risk.
Audit Metadata