Axiomata Skill Evaluator En
PassAudited by ClawScan on May 13, 2026.
Overview
The skill appears to be a purpose-aligned local evaluator that runs bundled Python scripts on user-selected skill folders, with some metadata/setup inconsistencies to notice.
This looks safe for its stated purpose if you are comfortable running bundled Python scripts that inspect local skill folders. Verify the Python environment, use only intended target paths, and do not provide any credentials unless the publisher gives a clear, purpose-aligned explanation.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent or user may run local Python code that reads the selected skill directory to generate evaluation results.
The skill expects local execution of bundled Python scripts against a user-selected skill path. This is central to the evaluator purpose and is disclosed, but users should recognize that installing the skill enables local script execution.
python3 ${EVAL_PATH}evaluator.py <SKILL_PATH> --verbose
python3 ${EVAL_PATH}eval-skill.py <SKILL_PATH> --verboseRun the scripts only from the installed skill directory and point them at skill folders you intend to inspect.
The skill may not run in an environment without Python and the needed Python modules, despite the metadata saying no requirements.
The metadata declares no runtime requirements, while SKILL.md documents python3 commands and the visible code imports modules such as yaml. This is a setup/declaration gap rather than evidence of malicious installation behavior.
Required binaries (all must exist): none ... No install spec — this is an instruction-only skill.
Publisher should declare Python/runtime requirements; users should verify the local Python environment before relying on the evaluator.