swarm-migrate

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands defined within YAML specification files located in the swarm-specs/ directory. These commands are used for codebase transformations (transform.command) and integrity verification (verify[].command) across different repository worktrees.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes the Agent tool to dispatch and manage parallel sub-agents. These sub-agents are responsible for performing git operations and executing the transformation logic in isolated environments, which is the core architectural design for the migration swarm.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with GitHub's infrastructure using the gh CLI to push code and manage pull requests. These network operations are strictly limited to the repositories specified in the user's migration plan.
  • [DATA_EXFILTRATION]: While the skill accesses and pushes code, it includes explicit instructions to quarantine credentials and avoid logging authentication tokens, focusing only on repository URLs and PR status reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 01:09 AM
Security Audit — agent-trust-hub — swarm-migrate