The Agent Skills Directory

[SAFE]: The skill provides a structured framework for converting git diffs and commit messages into educational documentation. It emphasizes integrity, trade-off analysis, and audience-appropriate explanation depth. No suspicious patterns such as obfuscation, credential harvesting, or unauthorized network activity were detected.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it analyzes external data sources like git diffs and commit messages. 1. Ingestion points: Data is pulled from repository history and pull requests during the EXTRACT phase. 2. Boundary markers: The skill mandates the use of '[Inference: evidence]' labels to distinguish interpreted content from facts. 3. Capability inventory: The skill's output is limited to markdown-based documentation and diagrams; it cannot execute code or access sensitive files. 4. Sanitization: It relies on instructional safeguards to prevent the inclusion of security-sensitive implementation details.

tome