grug-review
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data in the form of branch diffs, commits, and repository contents.
- Ingestion points: Reads code from git branches, commits, staged changes, and working directories.
- Boundary markers: None explicitly defined in the instructions.
- Capability inventory: Uses read-only git commands (
status,diff,logs,merge-base) and exploratory probes to navigate the codebase. - Sanitization: No explicit sanitization or escaping of external content is mentioned.
- [SAFE]: The skill contains no malicious patterns, obfuscation, or unauthorized network operations. Its instructions specifically remind the agent to remain read-only unless directed otherwise by the user.
Audit Metadata