understand-diff
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
git diffandgrepto identify changes and search metadata. This is standard behavior for a code analysis tool and is restricted to the local repository. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from code changes and knowledge graph files which could theoretically contain malicious instructions.
- Ingestion points:
SKILL.md(Steps 2, 4, and 5) instructs the agent to read output fromgit diffand search.understand-anything/knowledge-graph.json. - Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore instructions embedded within the files being analyzed.
- Capability inventory: The skill can execute shell commands (
git,grep) and perform file writes to.understand-anything/diff-overlay.json. - Sanitization: No sanitization or validation of the content retrieved from the diffs is specified before the agent processes it.
Audit Metadata