custom-quality-check

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to identify modified files and run linting tools. * Evidence: Invokes git diff --name-only and uv run ruff check --fix within SKILL.md.
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection via maliciously crafted filenames. * Ingestion points: File paths are ingested from the output of git diff in SKILL.md. * Boundary markers: None. The script does not utilize delimiters or specific instructions to treat filenames as literal data. * Capability inventory: The skill can execute subprocesses via shell interpolation and perform file write operations through the Ruff auto-fix feature. * Sanitization: None. The $CHANGED_PY variable is interpolated directly into a shell command without escaping or validation, which could allow for command execution if a filename contains shell metacharacters.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 03:27 PM