readme-docs

Warn

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's verification workflow (Step 8) instructs the agent to "Run every quick-start and example command that is safe and practical." This results in the execution of shell commands provided within the repository being analyzed, which could be malicious.
  • [REMOTE_CODE_EXECUTION]: The skill directs the agent to execute a specific test script (tests/test_readme_first_paragraph.py) found in the repository. Running scripts from an untrusted project carries a risk of arbitrary code execution.
  • [COMMAND_EXECUTION]: The skill requires running Markdown link and ASCII checks using tools likely found in the repository environment, which could be tampered with in a malicious project.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 30, 2026, 04:44 PM