research
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from primary sources (official docs, source code, specs) which could contain adversarial instructions that influence the agent's behavior.
- Ingestion points: External primary sources accessed during the background research task as instructed in
SKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or specific instruction-data isolation techniques when processing source content.
- Capability inventory: The agent is authorized to write to the repository to save research findings (
SKILL.md). - Sanitization: The skill does not mention explicit sanitization, filtering, or validation of the content retrieved from primary sources before it is processed or written to the file system.
Audit Metadata