agri-deep-research
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and process external full-text documents and research data. This creates a surface for Indirect Prompt Injection, where instructions hidden in source material could attempt to influence the agent's synthesis or behavior.
- Ingestion points: The skill performs a 'full-text-access first move' requesting access to the user's EndNote '.Data' folder and reference PDFs.
- Boundary markers: None explicitly defined in the provided file for isolating untrusted document content.
- Capability inventory: The skill executes a multi-agent pipeline and invokes citation verification scripts.
- Sanitization: No explicit sanitization or filtering of document content is mentioned before processing by the synthesis agents.
- [COMMAND_EXECUTION]: The skill references the execution of local scripts to perform verification tasks.
- Evidence: Use of 'scripts/verify_citations.py' as part of a four-gate citation verification process.
Audit Metadata