omc-teams
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Uses tmux and omc CLI commands to orchestrate parallel worker processes. This is the primary function of the skill and uses standard process management.
- [EXTERNAL_DOWNLOADS]: Recommends installation of official CLI tools from Anthropic, OpenAI, and Google. These are verified dependencies from well-known technology organizations.
- [DATA_EXFILTRATION]: Captures output from tmux panes using tmux capture-pane to verify process state and task progress. This allows the skill to verify that commands were correctly submitted.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by processing untrusted task descriptions that are then passed to downstream AI workers. 1. Ingestion points: User-provided task string in SKILL.md. 2. Boundary markers: Absent for the worker task input. 3. Capability inventory: Downstream CLI agents have capabilities for file system and network access. 4. Sanitization: Not explicitly performed within the skill instructions.
Audit Metadata