The Agent Skills Directory

[SAFE]: The skill consists of markdown templates, reference documentation, and a local validation script. No malicious patterns or security risks were identified.
[COMMAND_EXECUTION]: The skill includes scripts/check-output.py, which is a local script designed to validate that generated management deliverables meet minimum length and content requirements. It operates only on local file paths provided by the agent and performs no dangerous operations like network access, shell command execution, or data exfiltration.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user requests to generate management artifacts, it does not ingest external untrusted data from the web or other unverified sources, minimizing the risk of indirect prompt injection. The attack surface is limited to standard user-to-agent interactions.

team-management