explain-diff
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely within the local environment. It uses
git diffand filesystem read operations to gather context about code changes and writes the generated documentation to a local directory (defaulting to.explain/). - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from source code files and git diffs to generate its explanations. However, the risk is classified as safe/negligible because the generated output (HTML/Markdown) is intended for human reading rather than automated system execution.
- Ingestion points: Project source code files and the output of
git diffcommands (SKILL.md workflow). - Boundary markers: The instructions define a rigid structure for the output (Background, Intuition, Code, Quiz sections), which serves as a natural boundary.
- Capability inventory: Filesystem read (source code) and filesystem write (artifact creation and
.gitignoremodification). - Sanitization: The agent is directed to transform the input into a specific technical writing style, which involves synthesis rather than direct pass-through of potentially malicious instructions.
Audit Metadata