safe-change
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to find and execute local tests (
RUN: <test-cmd>) after code modification. This execution is scoped to the project environment and is necessary for the skill's primary function of verifying code changes. - [INDIRECT_PROMPT_INJECTION]: As the skill ingests and processes existing source code files to perform its tasks, it possesses an attack surface for indirect prompt injection. Malicious instructions could be hidden in comments or strings within the files the agent is instructed to modify.
- Ingestion points: Reads existing code files, module definitions, and test files (
SKILL.md). - Boundary markers: None specified for the code ingestion process.
- Capability inventory: File system read/write, shell command execution for testing, and invocation of secondary skills (
impact-check,verify-before-commit). - Sanitization: No explicit sanitization of input code content is described.
Audit Metadata