skills.rest
dhar174
dhar174COMMUNITY
Explore all skills by this author →
WebsiteGitHub

agentic-eval

Implement self-critique and evaluator-optimizer pipelines for iterative AI output refinement.

1|Updated Dec 22, 2025
Search Tags:#evaluation#reflection#agentic workflow#self-critique#iterative refinement#LLM-as-judge
One-click install
npx skills add https://github.com/dhar174/langgraph_system_generator --skill agentic-eval-dhar174
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentic-eval
Source: https://github.com/dhar174/langgraph_system_generator/tree/main/.github/skills/agentic-eval
Command: npx skills add https://github.com/dhar174/langgraph_system_generator --skill agentic-eval-dhar174

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables AI agents to iteratively improve their own outputs by incorporating self-evaluation and refinement loops, leading to higher quality and more reliable results.

Core Features & Use Cases

  • Self-Critique & Reflection: Agents can analyze their own work against defined criteria and identify areas for improvement.
  • Evaluator-Optimizer Pipelines: Decouples generation and evaluation for clearer workflows and targeted refinement.
  • Test-Driven Code Refinement: Specifically designed for code generation, using tests to drive bug fixing and quality improvements.
  • Use Case: An agent generates a complex report. This skill allows the agent to review the report for accuracy, clarity, and adherence to a style guide, then automatically revise it until it meets the required standards.

Quick Start

Use the agentic-eval skill to reflect and refine an AI-generated code snippet based on provided test cases.

Frequently Asked Questions about agentic-eval

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I implement self-critique for iterative refinement of AI agent outputs?▼

Self-critique for iterative refinement is implemented by having agents analyze their own work against defined criteria, identify areas for improvement, and automatically revise outputs until they meet required standards.

What is an evaluator-optimizer pipeline and when do I need it for LLM workflows?▼

An evaluator-optimizer pipeline decouples generation and evaluation into separate stages for clearer workflows and targeted iterative refinement. You need it when measuring and improving LLM response quality with structured feedback.

Can I use test-driven code refinement to fix AI-generated code snippets?▼

Yes, test-driven code refinement uses provided test cases to drive bug fixing and quality improvements in AI-generated code snippets, automatically iterating until the code passes the defined tests.

How does LLM-as-judge evaluation work for measuring agent response quality?▼

LLM-as-judge evaluation works by using a language model to review generated outputs against rubric-based criteria, providing structured feedback that enables the system to measure and improve overall agent response quality.

Do I need external dependencies to build an agentic workflow with reflection loops?▼

No external dependencies are required to build an agentic workflow with reflection loops. The skill provides self-contained patterns and techniques for adding self-evaluation and iterative improvement to your agents.

Related Skills

Explore More Software Engineering Skills→
Manavv007Manavv007

agentic-eval

Implement iterative evaluation and refinement loops for AI agent outputs.

Community
Advanced
iiammaeiiammae

sarif-parsing

Parse SARIF files with filtering, deduplication, and format conversion.

Community
Intermediate
lluppesmslluppesms

agentic-eval

Evaluate and refine AI agent outputs through self-critique and reflection loops.

Community
Advanced
ancolemanancoleman
503

evaluating-llms

Evaluate LLM systems with automated metrics, LLM-as-judge, and benchmarks.

Community
Advanced