The Agent Skills Directory

[SAFE] (SAFE): Analysis of the skill instructions and actions shows a focus on documentation tasks (reading specs, creating markdown, adding diagrams). No prompt injection, data exfiltration, or unauthorized command execution patterns were detected.
[INDIRECT_PROMPT_INJECTION] (INFO): The skill possesses an ingestion surface for untrusted data. 1. Ingestion Points: docs/spec.md, feature specifications, and ADR files. 2. Boundary Markers: Absent. 3. Capability Inventory: createFile, editFiles, and git commit. 4. Sanitization: Absent. Despite the lack of sanitization, the risk is negligible as the capabilities are restricted to generating static markdown documentation and diagrams within a controlled directory structure.

arc42-documentation