holistic-linting-resolver
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external Python source files and fetches documentation, which defines an indirect prompt injection surface. This behavior is necessary for its function as a linting resolver.
- Ingestion points: Source code read from local files and documentation retrieved via network requests.
- Boundary markers: Procedural steps and verification requirements act as implicit boundaries.
- Capability inventory: Shell command execution, file reading, and network fetching.
- Sanitization: No explicit sanitization of ingested code content is performed.
- [EXTERNAL_DOWNLOADS]: Fetches documentation from official domains for Mypy and Basedpyright to assist in error resolution.
- Evidence: URLs targeting mypy.readthedocs.io and docs.basedpyright.com.
- [COMMAND_EXECUTION]: Utilizes standard development tools and commands to verify fixes and locate environment-specific resources.
- Evidence: Execution of uv, ruff, mypy, and ripgrep via shell.
- [SAFE]: The skill implements strict workflows that prevent the use of suppression comments, encouraging genuine code fixes and maintaining project quality standards. All external tools and resources are standard for Python development environments.
Audit Metadata