research
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data during its research phase.
- Ingestion points: External primary sources, specifications, and APIs are accessed by background agents as described in SKILL.md.
- Boundary markers: The process utilizes a multi-agent structure where a separate 'Judge' agent evaluates researcher outputs, providing a layer of separation between data ingestion and final reporting.
- Capability inventory: The skill does not possess file-write or system execution capabilities beyond the research and reporting tasks.
- Sanitization: The 'Judge' is explicitly tasked with checking sources behind decisive claims and distinguishing independent corroboration from repeated source use.
- [NO_CODE]: The skill consists solely of natural language instructions and metadata. It does not contain scripts, shell commands, or external software dependencies, significantly reducing the technical attack surface.
Audit Metadata