run-research-pipeline
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and synthesize outputs from external LLMs (Claude, Gemini, GPT) provided by the user in Phase 3.
- Ingestion points: User input in Phase 1 via
$ARGUMENTSand external model responses pasted by the user in Phase 3. - Boundary markers: The skill uses XML-style tags (e.g.,
<model-response>,<research-pipeline-output>) to structure data, but does not include explicit instructions for the agent to ignore potential commands embedded within the untrusted external text. - Capability inventory: The skill orchestrates other tools/skills like
research-interviewerandcreate-research-briefto process the data. - Sanitization: There is no evidence of input validation or sanitization for the content provided from external models.
Audit Metadata