orchestrate
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a structured workflow coordination layer that does not introduce new security vulnerabilities.
- [COMMAND_EXECUTION]: Uses the Bash tool specifically for jq-based JSON validation of manifest files to ensure data integrity.
- [PROMPT_INJECTION]: Implements a manual checkpoint and automated validation of agent-generated plans to mitigate risks from malicious user input.
- [DATA_EXFILTRATION]: Accesses internal state files in localized project directories ('.claude/squad/runs/') with no outbound network activity.
Audit Metadata