The Agent Skills Directory

[SAFE]: The skill consists of Markdown documentation and templates with no executable scripts, binaries, or active code components.
[SAFE]: A strict role separation policy is enforced, preventing a single subagent from performing both implementation and validation, which serves as a security and quality control measure.
[SAFE]: The execution model is centered on a user-approved plan and requires no further user interaction for automated steps, but does not execute arbitrary code itself.
[SAFE]: No instances of prompt injection, data exfiltration, or hardcoded credentials were identified within the skill files.
[NO_CODE]: The skill does not include any functional software code or external package dependencies for its operation.

subagent-orchestration