content-research-writer
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources and user-provided drafts, creating an attack surface where embedded instructions could influence agent behavior.
- Ingestion points: External information retrieved during the research process (SKILL.md, 'Conduct Research' section) and user-supplied text sections provided for feedback (SKILL.md, 'Provide Section-by-Section Feedback' section).
- Boundary markers: Absent. The instructions do not specify the use of delimiters or warnings to ignore potentially malicious instructions embedded within the research material or drafts.
- Capability inventory: No high-risk capabilities such as arbitrary command execution, sensitive file access, or network exfiltration were found in the skill's instruction set.
- Sanitization: Absent. There is no logic for filtering or sanitizing the content fetched from external sources before it is processed by the agent.
Audit Metadata