The Agent Skills Directory

[SAFE]: The skill operates as a purely instructional guide for using a project-specific CLI tool (facts) to manage and refine documentation.
[COMMAND_EXECUTION]: The skill instructs the agent to use the facts utility to list, check, and edit project data. These operations are essential to the skill's purpose and are performed with explicit user oversight.
[SAFE]: No indicators of data exfiltration, unauthorized network access, or persistence mechanisms were found. The skill maintains a high safety profile by requiring user approval for all modifications.
[SAFE]: Indirect Prompt Injection Surface Analysis:
Ingestion points: Data is read from the project environment via facts list and facts list --tags "draft" (SKILL.md).
Boundary markers: Not explicitly implemented in the instructions.
Capability inventory: The skill can modify the project state via facts edit, facts add, and facts remove (SKILL.md).
Sanitization: No explicit sanitization is described, but the mandatory human-in-the-loop requirement for every change provides an effective security control.

facts-refine