pre-pr-checks
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various local development tools and system commands to perform project checks.
- Evidence: Use of
task,git,ruff,pytest,mypy,pyright, and language-specific runners for R (devtools,lintr,covr) and Julia (Pkg,Aqua). - The skill also performs file modifications and commits via
git commitfor formatting and linting fixes. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local environment which could potentially influence agent behavior.
- Ingestion points: The agent reads local source code files, git diff outputs, and the stdout/stderr of various test and linting tools (
SKILL.md). - Boundary markers: No specific boundary markers or instructions to ignore embedded instructions in the processed files are defined.
- Capability inventory: The skill has the capability to write to the file system and commit changes to the repository via subprocess calls (
SKILL.md). - Sanitization: No evidence of sanitization or filtering of the content read from files or tool outputs before they are processed by the agent.
Audit Metadata