# Pollack AI Lab

## Docs

- [ACP Spring Boot Autoconfiguration](https://lab.pollack.ai/docs/acp-java-sdk/autoconfig.md): Spring Boot autoconfiguration for the ACP Java SDK — auto-configured clients, agents, and transports with property-driven configuration.
- [ACP Java SDK](https://lab.pollack.ai/docs/acp-java-sdk/index.md): A Java SDK for the Agent Client Protocol — build both clients and agents that work with Zed, JetBrains, VS Code, and any ACP-compliant editor.
- [Java API Reference](https://lab.pollack.ai/docs/acp-java-sdk/reference/java.md): Complete API reference for the ACP Java SDK — client, agent, protocol types, transports, and test utilities.
- [01 first contact](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/01-first-contact.md)
- [02 protocol basics](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/02-protocol-basics.md)
- [03 sessions](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/03-sessions.md)
- [04 prompts](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/04-prompts.md)
- [05 streaming updates](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/05-streaming-updates.md)
- [06 update types](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/06-update-types.md)
- [07 agent requests](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/07-agent-requests.md)
- [08 permissions](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/08-permissions.md)
- [09 session resume](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/09-session-resume.md)
- [10 cancellation](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/10-cancellation.md)
- [11 error handling](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/11-error-handling.md)
- [12 echo agent](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/12-echo-agent.md)
- [13 agent handlers](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/13-agent-handlers.md)
- [14 sending updates](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/14-sending-updates.md)
- [15 agent requests](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/15-agent-requests.md)
- [16 in memory testing](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/16-in-memory-testing.md)
- [17 capability negotiation](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/17-capability-negotiation.md)
- [18 terminal operations](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/18-terminal-operations.md)
- [19 mcp servers](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/19-mcp-servers.md)
- [21 async client](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/21-async-client.md)
- [22 async agent](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/22-async-agent.md)
- [23 spring boot agent](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/23-spring-boot-agent.md)
- [24 spring boot client](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/24-spring-boot-client.md)
- [28 zed integration](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/28-zed-integration.md)
- [29 jetbrains integration](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/29-jetbrains-integration.md)
- [30 vscode integration](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/30-vscode-integration.md)
- [Tutorial](https://lab.pollack.ai/docs/acp-java-sdk/tutorial/index.md): A progressive tutorial for learning the ACP Java SDK — from client basics to IDE integration.
- [Agent Configuration](https://lab.pollack.ai/docs/agent-bench/agent-config.md): How to configure any CLI tool as an Agent Bench agent
- [CLI Reference](https://lab.pollack.ai/docs/agent-bench/cli-reference.md): All Agent Bench commands, flags, and usage patterns
- [Getting Started with Agent Bench](https://lab.pollack.ai/docs/agent-bench/getting-started.md): Test your AI coding agent against benchmarks in 5 minutes
- [Jury System](https://lab.pollack.ai/docs/agent-bench/jury-system.md): How Agent Bench grades agent output using cascaded tiers of judges
- [Defaults Philosophy](https://lab.pollack.ai/docs/agent-client/explanation/defaults-philosophy.md): LOOSE vs STRICT mode — how AgentClient manages default permissiveness across providers
- [Getting Started](https://lab.pollack.ai/docs/agent-client/howto/getting-started.md): Run your first agent task with Spring AI Agent Client in under 5 minutes
- [Structured Output](https://lab.pollack.ai/docs/agent-client/howto/structured-output.md): How to get structured JSON responses from agent tasks using JSON Schema
- [Switching Providers](https://lab.pollack.ai/docs/agent-client/howto/switching-providers.md): How to run the same agent task with Claude, Codex, or Gemini using Maven profiles
- [Claude Reference](https://lab.pollack.ai/docs/agent-client/reference/claude-reference.md): Complete configuration reference for the Claude Code agent provider
- [Codex Reference](https://lab.pollack.ai/docs/agent-client/reference/codex-reference.md): Complete configuration reference for the OpenAI Codex agent provider
- [Gemini Reference](https://lab.pollack.ai/docs/agent-client/reference/gemini-reference.md): Complete configuration reference for the Google Gemini agent provider
- [Portable Options Reference](https://lab.pollack.ai/docs/agent-client/reference/portable-options.md): AgentOptions interface, mode system, and the precedence rules that govern agent behavior across providers
- [Agent Sessions](https://lab.pollack.ai/docs/agent-client/reference/sessions.md): Persistent, multi-turn agent conversations with session lifecycle management
- [Lesson 1: Your First Agent Task](https://lab.pollack.ai/docs/agent-client/tutorial/01-first-task.md): Create a file using an agent — set up, run, and verify
- [Lesson 2: Multi-Provider](https://lab.pollack.ai/docs/agent-client/tutorial/02-multi-provider.md): Run the same task with Claude, Codex, and Gemini — the client code stays the same
- [Lesson 3: Read and Transform](https://lab.pollack.ai/docs/agent-client/tutorial/03-read-and-transform.md): Read files, perform data transformation, and write structured output
- [Lesson 4: Git Operations](https://lab.pollack.ai/docs/agent-client/tutorial/04-git-operations.md): Find lost git changes and merge them — agents working with version control
- [Agent Client Tutorial](https://lab.pollack.ai/docs/agent-client/tutorial/index.md): Learn Agent Client step by step — from your first task to git operations
- [API Reference](https://lab.pollack.ai/docs/agent-judge/api-reference.md): All public types, interfaces, and records in Agent Judge
- [Built-in Judges](https://lab.pollack.ai/docs/agent-judge/built-in-judges.md): Catalog of built-in judges across Agent Judge modules
- [Writing Custom Judges](https://lab.pollack.ai/docs/agent-judge/custom-judge.md): Build deterministic, LLM-powered, and RAG judges
- [Design Philosophy](https://lab.pollack.ai/docs/agent-judge/design-philosophy.md): Why Agent Judge is built the way it is — zero deps, functional interface, sealed scores, cascaded cost
- [Getting Started with Agent Judge](https://lab.pollack.ai/docs/agent-judge/getting-started.md): Add evaluation to your agent pipeline
- [Jury System](https://lab.pollack.ai/docs/agent-judge/jury-system.md): Combine judges with voting strategies, tiered cascades, and jury composition
- [Tutorial: Build an Evaluation Pipeline](https://lab.pollack.ai/docs/agent-judge/tutorial.md): Build an evaluation pipeline from a single judge to a multi-judge jury
- [AgentLoop](https://lab.pollack.ai/docs/agent-workflow/agent-loop.md): Ready-to-use SWE agent — tools, session memory, observability, and a simple run/chat API backed by Spring AI's agent loop
- [Annotation Model](https://lab.pollack.ai/docs/agent-workflow/annotation-model.md): Declare agents with @Agent, compose workflows inside AgentHandler, and wire exception handling with @ExceptionHandler and @AgentAdvice
- [Agent Workflow API Reference](https://lab.pollack.ai/docs/agent-workflow/api-reference.md): Step interface, Workflow builder, WorkflowGraph IR, gates, context, StepRunner, and TraceRecorder
- [DSL Primitives](https://lab.pollack.ai/docs/agent-workflow/choosing-a-pattern.md): 10+ composable primitives for building agentic pipelines — sequential, branch, loop, parallel, decision, gate, supervisor, and more
- [Durability](https://lab.pollack.ai/docs/agent-workflow/durability.md): Crash recovery, checkpointing, and distributed execution for agent workflows
- [Workflow DSL Examples](https://lab.pollack.ai/docs/agent-workflow/examples.md): Complete, runnable examples — validated with real LLM calls against GPT-4.1
- [Getting Started with Agent Workflow](https://lab.pollack.ai/docs/agent-workflow/getting-started.md): Compose steps into workflows with typed context, portable runtimes, and quality gates
- [Step Parameterization](https://lab.pollack.ai/docs/agent-workflow/parameterization.md): How to get data into and out of steps — constructor injection, input chaining, context keys, metadata publishing, and mixed patterns
- [Trace Capture](https://lab.pollack.ai/docs/agent-workflow/trace-capture.md): Capture per-step JSONL trace files and wire them through the workflow journal
- [Tutorial: Build a Workflow](https://lab.pollack.ai/docs/agent-workflow/tutorial.md): Build a multi-step AI workflow from a single step to a supervised agent pipeline
- [What's New](https://lab.pollack.ai/docs/agent-workflow/whats-new.md): Release highlights for Agent Workflow
- [Wiring Complex Pipelines](https://lab.pollack.ai/docs/agent-workflow/wiring.md): How to structure the constructor and Spring configuration for workflows with many collaborating steps
- [Your First Research Agent](https://lab.pollack.ai/docs/agento-studio/getting-started.md): Build a file-based research KB and teach an AI agent to navigate it — in 20 minutes
- [Architecture](https://lab.pollack.ai/docs/bud/architecture.md): Three modules, two protocols, zero API keys
- [Getting Started](https://lab.pollack.ai/docs/bud/getting-started.md): Install Bud and build your first Spring Boot project
- [Tools & Reference Projects](https://lab.pollack.ai/docs/bud/tools-reference.md): All 23 MCP tools and 9 reference projects
- [Agent Experiment API Reference](https://lab.pollack.ai/docs/experiment-driver/api-reference.md): Configuration, dataset format, invoker contract, and result model
- [Creating Experiments](https://lab.pollack.ai/docs/experiment-driver/creating-experiments.md): Design datasets, define variant ladders, filter items, and analyze results
- [Diagnostic Reasoning](https://lab.pollack.ai/docs/experiment-driver/diagnostic-reasoning.md): Classify failure gaps, generate remediation actions, and feed the improvement flywheel
- [Getting Started with Agent Experiment](https://lab.pollack.ai/docs/experiment-driver/getting-started.md): Run your first AI agent evaluation: dataset, agent, jury, and variant comparison
- [Building a Jury](https://lab.pollack.ai/docs/experiment-driver/jury-system.md): Three-tier cascaded evaluation: deterministic, structural, and semantic judges
- [Pipeline](https://lab.pollack.ai/docs/experiment-driver/pipeline.md): Three-phase orchestration: analyze a project, generate a plan, then execute with enriched context
- [Sessions and Sweeps](https://lab.pollack.ai/docs/experiment-driver/sessions-and-sweeps.md): Group variant results into sessions and coordinate multi-session sweeps
- [Loopy CLI Reference](https://lab.pollack.ai/docs/loopy/cli-reference.md): All flags, slash commands, execution modes, and configuration options
- [Extending Loopy](https://lab.pollack.ai/docs/loopy/extending.md): Custom skills, subagents, tool profiles, listeners, and the programmatic API
- [Getting Started with Loopy](https://lab.pollack.ai/docs/loopy/getting-started.md): Install, configure, and run your first agent session in under 5 minutes
- [Migration: spring-ai-community to markpollack](https://lab.pollack.ai/docs/migration/spring-ai-community-to-markpollack.md): Coordinate and repository changes for projects that moved from the spring-ai-community GitHub organization
- [Code Coverage v1 — Knowledge Injection Baseline](https://lab.pollack.ai/experiments/code-coverage-v1.md): 9 variants testing progressive knowledge injection on Spring Boot test generation
- [Code Coverage v2 — Skills vs Knowledge Bases](https://lab.pollack.ai/experiments/code-coverage-v2.md): 7 variants on Spring PetClinic testing whether structured skills outperform flat knowledge injection
- [Code Coverage v3 — The Exemplar Effect](https://lab.pollack.ai/experiments/code-coverage-v3.md): When existing tests use older patterns, skills can't override them. The codebase is the agent's primary teacher.
- [Experiments](https://lab.pollack.ai/experiments/index.md): Controlled studies measuring what moves the needle for AI agent reliability
- [Issue Classification — Infrastructure vs Prompts](https://lab.pollack.ai/experiments/issue-classification.md): SWE-bench Lite: does infrastructure-level optimization beat prompt engineering?
- [Pollack AI Lab](https://lab.pollack.ai/index.md): Tools and experiments for building agents that work — and understanding why they work
- [Forge Methodology](https://lab.pollack.ai/methodology/forge.md): Deterministic customization pipeline — from knowledge to structured agent execution
- [Four-Tier Jury](https://lab.pollack.ai/methodology/four-tier-jury.md): Cascaded evaluation — deterministic first, LLM last
- [Improvement Flywheel](https://lab.pollack.ai/methodology/improvement-flywheel.md): Loss-driven iteration — from measured behavioral gaps to targeted interventions
- [Methodology](https://lab.pollack.ai/methodology/index.md): How we design experiments, evaluate agents, and analyze behavior
- [Knowledge Base Design](https://lab.pollack.ai/methodology/knowledge-base-design.md): How to structure domain knowledge for agent consumption — routing tables, progressive disclosure, and the Diataxis weighting
- [Knowledge Base Freshness](https://lab.pollack.ai/methodology/knowledge-base-freshness.md): How knowledge stays true after it's written — cached routing judgments, two-channel freshness, and rituals that consume drift signals
- [Knowledge-Directed Execution](https://lab.pollack.ai/methodology/knowledge-directed-execution.md): The thesis: knowledge + structured execution > model
- [Markov Fingerprinting](https://lab.pollack.ai/methodology/markov-fingerprinting.md): Modeling agent tool-call traces as stochastic processes to reveal behavioral patterns
- [Structured Agent Execution (SAE)](https://lab.pollack.ai/methodology/structured-agent-execution.md): Phased execution with checkpoints, guard rails, and deterministic preprocessing
- [ACP Java SDK](https://lab.pollack.ai/projects/acp-java-sdk.md): Agent Communication Protocol — build agents, consume agents, integrate with IDEs using a standard protocol
- [Agent Bench](https://lab.pollack.ai/projects/agent-bench.md): Benchmarking suite for Java-centric AI agents on real-world software engineering tasks
- [Agent Client](https://lab.pollack.ai/projects/agent-client.md): Portable Java API for autonomous CLI agents — Claude Code, Codex, Gemini, Amazon Q, and Amp
- [Agent Experiment](https://lab.pollack.ai/projects/agent-experiment.md): End-to-end experiment driver for evaluating AI coding agents against fixture datasets
- [Agent Hooks](https://lab.pollack.ai/projects/agent-hooks.md): Portable hook API for steering agent behavior at the tool-call boundary — write once, run on any runtime
- [Agent Journal](https://lab.pollack.ai/projects/agent-journal.md): Structured event logging, behavioral trace extraction, and human feedback for agent research
- [Agent Judge](https://lab.pollack.ai/projects/agent-judge.md): Framework-neutral evaluation layer for AI agent output across Spring AI, LangChain4j, Koog, and CLI agents
- [Agent Memory](https://lab.pollack.ai/projects/agent-memory.md): Progressive memory management for Spring AI — from context compaction to autonomous memory control
- [Agent Sandbox](https://lab.pollack.ai/projects/agent-sandbox.md): Isolated command execution — Local, Docker, and E2B cloud backends behind a unified API
- [Agent Workflow](https://lab.pollack.ai/projects/agent-workflow.md): Build agents that work — and measure why they work. Multi-step pipelines with typed context, quality gates, and portable runtimes.
- [Agento Studio](https://lab.pollack.ai/projects/agento-studio.md): A systematic approach to growing AI agents — judges tell you if the agent worked, journals tell you why
- [Agento University](https://lab.pollack.ai/projects/agento-university.md): Hierarchical multi-agent platform for continuous, autonomous management of software development projects
- [AgentWorks BOM](https://lab.pollack.ai/projects/agentworks-bom.md): Bill of Materials for coordinated version management across all AgentWorks projects
- [Bud](https://lab.pollack.ai/projects/bud.md): ACP agent for Spring Boot development — proven patterns as a starting point, AI to adapt them
- [Claude Agent SDK (Java)](https://lab.pollack.ai/projects/claude-agent-sdk.md): Java SDK for Claude Code CLI integration — three-API architecture, sessions, MCP, multi-agent orchestration
- [Projects](https://lab.pollack.ai/projects/index.md): Tools for building, evaluating, and observing AI agents on the JVM
- [Loopy](https://lab.pollack.ai/projects/loopy.md): Loop-driven interactive coding agent CLI for Java developers
- [What's New](https://lab.pollack.ai/whats-new.md): Weekly release highlights across all Pollack AI Lab projects