review-elixir
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses git to identify changes and mix to run formatters, linters, and tests. These are standard, expected behaviors for an Elixir development tool.\n- [INDIRECT_PROMPT_INJECTION]: The skill reads external Elixir source files which could contain malicious instructions. Ingestion points: Elixir source files (.ex, .exs, .heex) and configuration files (mix.exs). Boundary markers: The Hard gates and Evidence gate require the agent to verify findings against the source code context. Capability inventory: Execution of mix and git. Sanitization: None.
Audit Metadata