citation-validator
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes research reports provided by users, which serves as an ingestion point for untrusted data. Maliciously crafted reports could contain instructions intended to exploit the agent's capabilities.
- Ingestion points: The agent is instructed to scan research content for factual claims (Step 1, instructions.md).
- Boundary markers: There are no instructions to use delimiters or to ignore commands embedded within the content being analyzed.
- Capability inventory: The skill employs WebSearch, WebFetch, and Read/Write tools for network and file system operations (SKILL.md).
- Sanitization: The instructions do not include procedures for sanitizing claims or external data before they are used in tool calls or written to the file system.
Audit Metadata