The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes Python scripts (validate_novelty_pack.py and test_validate_novelty_pack.py) that use subprocess.run to execute internal validation tools. Specifically, it invokes a sibling skill's validator (validate_review_pack.py) to ensure literature consistency.
Evidence: scripts/validate_novelty_pack.py calls subprocess.run(command, text=True, capture_output=True) where the command is constructed from local file paths and the Python executable.
[SAFE]: The SKILL.md file defines a clear protocol for research integrity, emphasizing skeptical review and adversarial questioning. It references sibling skills (research-paper-review, research-systematic-literature-review) for modular data handling.
[SAFE]: No hardcoded credentials, malicious obfuscation, or unauthorized network calls were detected. All URL references are internal to the research framework or standard markdown links.
[SAFE]: The project follows standard software engineering practices, including unit tests and schema-based validation for its output artifacts.

research-novelty-review