z-lint
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell-based linting and formatting tools (e.g., Biome, ESLint, Prettier, Ruff, Black, Go, Rust) and Git commands (
git diff,git ls-files). This is the intended primary function of the skill. - [INDIRECT_PROMPT_INJECTION]: The skill reads project configuration files, scripts, and lockfiles to determine its behavior, which constitutes an ingestion point for untrusted data from the repository.
- Ingestion points: Reads repository files including project instructions, wrappers, scripts, lockfiles, and package configurations (SKILL.md).
- Boundary markers: None explicitly defined for prompt delimiters, though it instructs the agent to "Match the package manager and installed versions" and "do not let a convenience launcher download a missing tool silently."
- Capability inventory: Execution of shell commands (linting/formatting tools) on project files based on gathered configuration.
- Sanitization: Not explicitly defined, though it instructs to "Preserve spaces and other filename characters when passing arguments" during file selection.
Audit Metadata