git-commit-summarizer

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection (HIGH): The skill ingests commit messages from git history, which are considered untrusted external data. These messages are interpolated directly into markdown reports without sanitization or protective delimiters. Evidence: Ingestion point at git_commit_analyzer.py (line 74), output generation in report_generator.py (line 89). The skill possesses file-write capabilities (report_generator.py, line 170) and command execution capabilities (git_commit_analyzer.py, line 60), creating a high-risk surface for downstream agent exploitation.
  • Command Execution (LOW): The skill uses subprocess.run to execute git commands. While it avoids shell injection by using a list of arguments, the usernames input is used to construct the --author flag without validation (line 61), which is a best-practice violation and could potentially lead to argument manipulation within the git command context.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 01:00 PM