skills/autumnsgrove/lattice/git-hooks/Gen Agent Trust Hub

git-hooks

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEPERSISTENCEPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [PERSISTENCE]: The skill installs Git hooks such as commit-msg, pre-commit, and pre-push. These are persistent automated scripts that run on the system whenever specific Git actions occur.
  • [PRIVILEGE_ESCALATION]: The instructions command the agent to apply chmod +x to files in the .git/hooks/ directory. This elevates the permission level of these files to allow their execution.
  • [COMMAND_EXECUTION]: The skill executes shell commands and utilizes uv run to trigger development tools like black and ruff during the hook execution process.
  • [DYNAMIC_EXECUTION]: The skill encourages the execution of an interactive installer script (install_hooks.sh) and various shell scripts copied into the Git configuration directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content, including Git commit messages and file data, which creates a surface for indirect prompt injection.
  • Ingestion points: Commit messages processed by the commit-msg hook and staged file content processed by the Python pre-commit hooks.
  • Boundary markers: The skill does not define specific markers or delimiters to protect against malicious instructions embedded in the processed data.
  • Capability inventory: Shell execution, Python tool execution via uv, and Git metadata access.
  • Sanitization: There is no evidence of sanitization or filtering of commit message content or file data before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 01:31 AM