Group: GitHub Markpollack

Sort by:Popular▼

✓Popular✓Newest

Memory store abstraction, token-budgeted retrieval, and LLM-powered compaction

Java SDK for interacting with mini-SWE-agent CLI - Low-level binding for autonomous software engineering

Journal integration for agent-workflow — records per-step events into agent-journal runs

Reusable tools for AI agents: file operations, shell execution, search

Bill of Materials for the Agent Suite

Full CLI entry point (BenchApp) and LLM judge (TestQualityJudge) for Agent Bench

Benchmark for Java focused Agents to determine how well they solve a range of problems

Autonomous CLI Agent Integrations for Spring AI

Reproducible Java experiment lifecycle for evaluating AI agents

Claude Agent SDK adapter for agent-hooks — bridges AgentHookProvider to Claude CLI hooks

Portable hook API for steering agent behavior at the tool-call boundary

Execution ledger — Experiment/Run tracking for agent workflows

AgentClient CLI-agent integration for Agent Judge - bridge from CLI-delegated agent execution to JudgmentContext

File comparison and semantic diff judges for agent evaluation

JetBrains Koog agent integration for Agent Judge - bridge from Koog agent execution to JudgmentContext

LangChain4j integration for Agent Judge - bridge from LangChain4j Result to JudgmentContext

Framework-neutral, status-first evaluation for AI agent outcomes

RAG evaluation judges for Agent Judge - faithfulness, contextual relevance, and hallucination detection

Spring AI ChatResponse evaluation bridge for Agent Judge - evaluated-side bridge, not judging-side LLM dependency

Core JBang launcher functionality for Spring AI Agents - AgentSpec, RunSpec, LocalConfigLoader