researchers-verifier
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources, creating a surface for indirect prompt injection. 1. Ingestion points: The skill uses WebFetch to retrieve content from external URLs and reads local files including SOURCES.md, track files, and {overrides}/research-preferences.md (SKILL.md, checklists.md). 2. Boundary markers: There are no instructions defining clear boundaries or delimiters to help the agent distinguish between data to be verified and potential instructions within that data (SKILL.md). 3. Capability inventory: The agent has access to powerful tools including WebFetch, WebSearch, Write, and Edit, which could be abused if the agent is manipulated by injected content (SKILL.md). 4. Sanitization: The instructions do not specify any validation, filtering, or sanitization of the content retrieved from external or local sources before processing.
Audit Metadata