git

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on Bash to perform Git operations such as adding, committing, and pushing changes. This gives the agent direct access to execute shell commands within the repository environment.
  • [PROMPT_INJECTION]: There is an indirect prompt injection surface (Category 8) because the skill analyzes repository context and Git state. 1. Ingestion points: Read operations on repository context and Git state (e.g., file contents, commit history). 2. Boundary markers: Absent. The instructions do not define clear delimiters for untrusted repository data. 3. Capability inventory: Subprocess execution via Bash for Git commands and file system access via Read. 4. Sanitization: The skill mentions validation but lacks specific mechanisms to sanitize repository data before it is processed by the AI or used in commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:06 AM