The Agent Skills Directory

[SAFE]: The skill logic is entirely instructional and focused on developer workflow management. No suspicious commands or behaviors were detected.
[DATA_EXPOSURE]: The skill references standard security practices, such as storing sensitive keys like JWT_SECRET in .env files rather than hardcoding them or including them in session handoffs. This is a recommended practice.
[PROMPT_INJECTION]: The skill processes local handoff.md files which represents an indirect prompt injection surface.
Ingestion points: The agent reads handoff.md or files in .handoff/ at the start of a session.
Boundary markers: The skill instructs the agent to read the file and confirm the next step with the user via a specific confirmation sentence.
Capability inventory: The skill does not define or use any dangerous tools or system-level capabilities.
Sanitization: The skill relies on natural language processing of text files and includes a manual confirmation step from the user before execution.

handoff