The Agent Skills Directory

[SAFE]: No security issues were detected across the analyzed files.
[PROMPT_INJECTION]: No direct instructions to bypass safety protocols or override agent behavior were found.
Evidence: Instructions in SKILL.md are strictly task-oriented and focus on taxonomy structure and quality.
[DATA_EXFILTRATION]: No network operations or sensitive data access patterns were identified.
Evidence: The Python script in scripts/run.py does not use networking libraries and processes only local workspace files such as papers/core_set.csv and queries.md.
[COMMAND_EXECUTION]: No shell command execution or dynamic code evaluation (eval/exec) was found.
Evidence: The script uses standard CSV and file I/O libraries to generate structured YAML data.
[REMOTE_CODE_EXECUTION]: No external downloads or remote code execution patterns are present.
Evidence: All dependencies are local to the repository environment.

taxonomy-builder