The Agent Skills Directory

[SAFE]: The skill is designed with a high degree of security awareness. It explicitly instructs the agent to treat repository-controlled content (commits, refs, paths) as untrusted data and to ignore any embedded instructions that might attempt to override safety controls or expand scope.
[COMMAND_EXECUTION]: While the skill involves performing Git operations, it provides specific security guidelines to prevent command injection, such as using the '--' separator and avoiding shell interpolation. It also warns against the automatic execution of repository hooks and other Git-specific extension points.
[DATA_EXFILTRATION]: The skill emphasizes the use of minimum credentials and explicitly prohibits embedding tokens in command lines or configuration, reducing the risk of accidental credential exposure during the migration process.

migrate-repository