skills/swm8023/skills/git-workflow/Gen Agent Trust Hub

git-workflow

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill manages the repository lifecycle by executing Git commands through the system shell.
  • Evidence: The skill description and lifecycle phases (prepare, checkpoint, finalize) involve executing git status, git push, git merge, and worktree management commands.
  • [INDIRECT_PROMPT_INJECTION]: The skill reads behavioral instructions from a local documentation file, which creates a surface where external data can influence the agent's logic.
  • Ingestion points: Configuration is read from docs/user/git-preferences.md during the prepare phase.
  • Boundary markers: Absent. The skill does not specify the use of delimiters or 'ignore' instructions when parsing the configuration file.
  • Capability inventory: The skill can modify the local file system, execute shell commands via the Git CLI, and perform network operations via git push.
  • Sanitization: Absent. The instructions do not mention validating or sanitizing the content of the preferences file before adopting the decisions defined within it.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 01:36 AM
Security Audit — agent-trust-hub — git-workflow