researchers-tech
Pass
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) because its core function involves processing untrusted external data.
- Ingestion points: Untrusted data enters the agent context via the
WebFetchandWebSearchtools while researching technical archives, mailing lists, and developer blogs (documented inSKILL.mdandPROJECT_RESEARCH.md). - Boundary markers: The instructions lack explicit boundary markers or systemic warnings to the agent to disregard instructions potentially embedded within the research material.
- Capability inventory: The skill allows the agent to use
EditandWritetools. While intended for documenting findings, an indirect injection from a malicious source could attempt to trick the agent into performing unauthorized file modifications. - Sanitization: There is no evidence of sanitization or validation of the content retrieved from external URLs before it is processed by the model.
Audit Metadata