interview-retro
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data including interview transcripts, job descriptions, and resumes, which are treated as untrusted input from potentially external sources.
- Ingestion points: Data enters the agent context via direct user input or via the
lark-doctool when reading external links (SKILL.md, Phase 0). - Boundary markers: The skill lacks explicit instructions to treat external text as data only or to ignore embedded instructions; no delimiters or warning markers are present in the processing logic.
- Capability inventory: The skill explicitly allows the
Bashtool in its frontmatter and performs write operations to external services via thelark-docskill to create new documents (SKILL.md, Phase 4). - Sanitization: There is no evidence of input filtering, escaping, or validation of the ingested external content before it is processed by the agent.
Audit Metadata