
Agentic eval
SkillGuides agents through self-critique, reflection loops and LLM-as-judge evaluation
- Open source
- 41.3K GitHub stars
- AI models and LLM tooling


Guides agents through self-critique, reflection loops and LLM-as-judge evaluation

Guides building, running and validating LLM evaluators with the Phoenix framework

Evaluate and iteratively improve AI agents on Google Cloud with an eval flywheel

Guides an agent through drafting, testing and tuning new agent skills with measured evals

Creates Arize LLM-as-judge evaluators and runs them on traces or experiment results

Build custom evaluation metrics that grade your AI agent the way you do

Guides agents through debugging LLM apps with Phoenix traces, datasets and experiments

Run local evals on Hugging Face models with inspect-ai or lighteval and pick a backend

Lets an agent create, list, update and delete Arize AI LLM provider integrations

Runs Expo skills through trigger, code and device screenshot checks

Guides an agent to build an llms.txt index file for a repository from its structure

Builds a minimal, vetted set of agent skills for a multi-step goal, with install checks
The closest alternatives to Eval driven dev are Agentic eval, Phoenix evals, Agent platform eval flywheel, Skill creator and Arize evaluator.
Yes. Agentic eval, Phoenix evals, Agent platform eval flywheel have a free or open-source option.