researchers-gov
Pass
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill possesses a surface for indirect prompt injection by combining external data ingestion with file-system modification tools.
- Ingestion points: Untrusted data enters via
WebFetchfrom various government agency websites and the Wayback Machine (e.g.,justice.gov,fbi.gov,sec.gov,web.archive.org). - Boundary markers: Absent. The skill instructions do not explicitly define delimiters for fetched content or warn the agent to ignore instructions embedded within the source documents.
- Capability inventory: The skill is granted
Read,Write,Edit,Grep,Glob,WebFetch, andWebSearchtools, allowing it to modify files based on findings. - Sanitization: Absent. The skill extracts specific fields (quotes, numbers, timelines) from raw text without explicit sanitization or validation of the fetched content.
Audit Metadata