ForkProbe eliminates the guesswork in choosing AI skills. It recommends a set of candidate skills (including a baseline), executes the same task across all candidates in parallel, and generates a comprehensive HTML report for side-by-side comparison. It supports both text-based outputs (like academic writing) and artifact-based outputs (like PPTX files, scientific figures, and research reports).
Use this skill when you are unsure which specialized skill will produce the best result, when comparing 2+ skills for a high-stakes task, or when you need to audit the difference between a baseline LLM and an optimized skill pipeline.
compare.py for text, figure_artifact.py for visuals, research_artifact.py for reports, and recommend.py for candidate discovery.Optimized for Claude Code and Codex.
This skill has not been reviewed by our automated audit pipeline yet.