harden
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the execution of shell commands such as 'bun run test:branch-cov', 'zig build', and 'git worktree'. These are part of the intended code review and testing workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill operates on repository source code and git diffs, which are external data sources that could theoretically contain adversarial content.
- Ingestion points: Source code files and git diffs from the OpenTUI repository.
- Boundary markers: No explicit delimiters are used to separate code content from instructions.
- Capability inventory: Includes file modification and shell command execution within the workspace.
- Sanitization: Data from the repository is processed without additional filtering or sanitization.
Audit Metadata