git-guardrails-claude-code
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a shell script (
scripts/block-dangerous-git.sh) designed to run as a PreToolUse hook. This script is intended to be executed locally to validate Git commands before they are processed by the agent. - [INDIRECT_PROMPT_INJECTION]: The safety script processes tool input data to identify dangerous command patterns.
- Ingestion points: The
scripts/block-dangerous-git.shscript reads JSON input from standard input representing the tool call context. - Boundary markers: Not applicable, as the script performs logical checks via grep rather than prompt interpolation.
- Capability inventory: The script utilizes
jqfor JSON parsing andgrepfor regex pattern matching to identify blocked commands. - Sanitization: The script uses
jqto extract specific fields from the input, which helps ensure that only the intended command string is evaluated.
Audit Metadata