Evaluate AI pipeline outputs using a weighted rubric and log iteration results.
npx skills add https://github.com/ctoscano/cw-hackathon-3 --skill evaluate-ctoscano
Please help me install this Agent Skill. Skill: evaluate Source: https://github.com/ctoscano/cw-hackathon-3/tree/main/.claude/skills/evaluate Command: npx skills add https://github.com/ctoscano/cw-hackathon-3 --skill evaluate-ctoscano
This Skill helps you automatically audit AI pipeline outputs and propose targeted, iterative improvements to prompts and configurations, enabling more consistent quality across pipelines such as DAP notes and intake questionnaires.
Run a full evaluation for the DAP notes pipeline with /evaluate run dap. You can also compare versions with /evaluate review dap.
High-intent search queries and answers about installing and using this skill.
You can evaluate AI pipeline outputs automatically by applying a weighted rubric to the generated results. This process audits outputs from pipelines like DAP notes and intake questionnaires to propose targeted iterative improvements.
The best way to iteratively improve AI prompts is to log evaluation results from a weighted rubric across runs. Tracking this iteration history guides subsequent edits to your configurations for consistent quality.
Yes, you can use a weighted rubric to audit DAP notes and intake questionnaire pipelines. The evaluation process loads pipeline references, identifies latest outputs, and applies expert personas to assess quality.
To run a full evaluation for the DAP notes pipeline, execute the run command. This triggers the automated evaluation process, which reads references and logs iteration results to guide prompt edits.
To compare versions when evaluating AI pipeline outputs, use the review command. This allows you to review previous iteration logs and evaluation results to track improvements across pipeline runs.
Yes, the evaluation process supports multiple AI pipelines by loading pipeline references and identifying latest outputs. It applies across DAP notes, intake questionnaires, and any future pipelines using the rubric.
selcukyucelGenerate structured evaluation datasets with binary rubric scoring for AI outputs.
skyasu2Extract actionable issues from pending code reviews and apply suggested edits.
appwriteAutomate deployment and configuration of serverless functions in Appwrite environments.
Devsoul2026Optimize and deploy LLM inference on NVIDIA GPUs with TensorRT-LLM.
