The Agent Skills Directory

[SAFE]: No security issues were identified. The skill's operations are confined to the local project directory and standard development commands.
[DATA_EXFILTRATION]: The skill reads local documents from the backlog and memory notes to provide status updates. No network tools or data exfiltration mechanisms are utilized.
[PROMPT_INJECTION]: Ingestion of project markdown files presents a surface for indirect prompt injection. However, the skill's capabilities are limited to informational reporting and git status checks. Ingestion points: backlog documents and memory files. Boundary markers: absent. Capability inventory: file listing and git status/branch/checkout. Sanitization: absent.

prime-feat