ln-12-delivery-reviewer
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to execute repository-defined commands, including build scripts, linters, type checkers, and test suites, to verify the integrity of code changes.
- [DYNAMIC_EXECUTION]: The framework leverages subagents to perform parallel, multi-perspective reviews (using the 'Six Thinking Hats' methodology), which involves spawning new execution contexts for specialized analysis.
- [INDIRECT_PROMPT_INJECTION]: As the skill ingests and analyzes external repository content and diffs, it is theoretically susceptible to instructions embedded in the code being reviewed; however, it implements scoping rules to focus specifically on causal paths and change deltas.
Audit Metadata