raycast-orchestrator
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to perform dynamic file loading by reading sub-skill definitions from the filesystem at runtime. It constructs file paths using variable spoke names (e.g.,
~/.agents/skill-clusters/skills/<spoke-name>/SKILL.md) and instructs the agent to load these files on demand. While the skill attempts to limit this behavior to a hardcoded list of 'spokes', the pattern of accessing the local filesystem—particularly the home directory—via dynamically generated paths is a sensitive capability that requires careful oversight.
Audit Metadata