sciomc
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not perform any unauthorized network operations, credential harvesting, or dangerous system modifications. Its primary function is the orchestration of sub-agents using an internal task execution mechanism.
- [PROMPT_INJECTION]: The skill processes user-defined research goals and findings generated by sub-agents. While this presents a surface for indirect prompt injection (where malicious content in researched files could attempt to influence the agent), the skill employs structured tags (e.g., [FINDING], [EVIDENCE]) and regex-based extraction to parse results, which provides a layer of data sanitization.
- [COMMAND_EXECUTION]: File system interactions are limited to the
.omc/research/directory for persisting session state and reports. This behavior is consistent with the skill's documented purpose of managing research sessions.
Audit Metadata