synthesize-research
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to ingest and analyze data from external, potentially untrusted sources such as customer support tickets, survey responses, and user interview transcripts.
- Ingestion points: Data enters the agent's context through pasted text, file uploads, and connected feedback sources as described in the Workflow section of SKILL.md.
- Boundary markers: The instructions do not explicitly specify the use of delimiters or specific 'ignore' directives for the ingested content, which is a common practice for qualitative research synthesis.
- Capability inventory: The skill consists entirely of markdown instructions and does not define any shell commands, network operations, file-writing tools, or other executable capabilities.
- Sanitization: The skill relies on standard natural language processing and does not include specific filtering or sanitization steps for the research text it processes.
Audit Metadata