CtrlK
BlogDocsLog inGet started
Tessl Logo

gamussa/langchain4j-agentic

Build and demo Java AI agent systems with langchain4j-agentic: workflow patterns, supervisor, custom Planner strategies (incl. the flagship typed-verdict / CriticResult-style critic pattern), plus MCP tools, A2A remote agents, build setup, and conference-demo storylines. Pinned to 1.15.0 / 1.15.0-beta25.

84

4.76x
Quality

89%

Does it follow best practices?

Impact

100%

4.76x

Average score across 2 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

task.mdevals/scenario-1/

Parallel Article Quality Pipeline

Problem Description

A media analytics company publishes several hundred articles per day across multiple channels and needs an automated quality-assurance pipeline to keep standards high without bottlenecking editorial staff. Each incoming article must be classified for sentiment and assigned a quality score, and those that score below an editorial threshold need to go through a refinement loop: an editor agent rewrites the article, the scorer re-evaluates it, and the cycle continues until the score is acceptable or a maximum number of revision attempts is exhausted.

Speed matters: the classification and initial scoring phase should process a whole batch of articles at once rather than one at a time down a serial queue. The engineering team wants to build this pipeline with the LangChain4j agentic framework in Java. They've been burned before by thread-safety bugs, runaway token bills, and refinement loops that never terminate, so reliability under load is a stated priority.

Output Specification

Produce a self-contained Java project with the following files:

  • build.gradle.kts — Gradle build configuration
  • src/main/java/pipeline/ — Java source files including:
    • Agent interfaces for article classification, quality scoring, and article refinement
    • A main/wiring class Pipeline.java that:
      • Constructs the parallel processing stage (processing multiple articles concurrently)
      • Constructs the iterative refinement loop stage
      • Chains these into a complete pipeline
      • Demonstrates the pipeline on a hardcoded sample batch of 3 articles (write them inline in the main method)
      • Produces an execution report after the run

The project must compile cleanly. The execution report file should be written to the project directory so it is visible after running.

README.md

tile.json