content-research
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of markdown-based instructions for research and fact-checking workflows. It does not perform network operations, file system modifications, or command execution. The reference to 'creator-style.md' is a local file reference for internal style consistency and is not a security risk. \n- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to gather and process external information, which creates a surface for indirect prompt injection. \n
- Ingestion points: Step 2 ('Gather Primary Evidence') involves reading external documentation, source code, first-party announcements, and other web content. \n
- Boundary markers: The 'Claim Ledger' table in Step 3 provides a logical structure for data but does not include technical delimiters or safety instructions to isolate untrusted content. \n
- Capability inventory: The skill files contain no executable code, subprocess calls, network-active scripts, or file-writing tools. \n
- Sanitization: No explicit sanitization or filtering logic is defined for the external research material.
Audit Metadata