The Agent Skills Directory

[SAFE]: The skill body consists entirely of natural language instructions for routing tasks. It does not contain any shell commands, scripts, or executable code.
[SAFE]: The skill accesses local documentation files (docs/agents/index.md and docs/agents/work-item-format.md) which is consistent with its stated purpose of following governance procedures. No external data is fetched or exfiltrated.
[SAFE]: The skill explicitly defines a scope limited to routing and reasoning, stating that it does not create issues, pull requests, or project items, which adheres to the principle of least privilege.

work-item-router