Skip to content

Integrations Overview

ACE provides both runners and step-based integrations for popular agentic frameworks. Some integrations are full runners, while others are composable pipeline steps you can drop into a custom Pipeline.

Available Integrations

Runner Framework Input Insight Level
ACELiteLLM LiteLLM (100+ providers) Questions Micro
LangChain LangChain Runnables Chain inputs Meso
BrowserUse browser-use Task strings Meso
ClaudeCode Claude Code CLI Task strings Meso
Claude SDK Anthropic Python SDK Task strings or ACESample Meso
OpenClaw OpenClaw transcripts JSONL trace files Meso
MCP Server MCP (stdio) Tool calls Micro
MCP Client Setup Claude Code, Cursor, Windsurf — Setup Guide
Opik Opik observability — Monitoring
Tracing Kayba tracing SDK @trace / start_span Cloud
Hosted API Kayba hosted API Trace files Cloud

The Pattern

All integration runners follow the same three-step pattern:

1. INJECT   — Add skillbook strategies to the agent's context
2. EXECUTE  — Run the external agent normally
3. LEARN    — Reflector + SkillManager update the skillbook

Quick Construction

Every runner offers a from_model() factory that builds ACE roles automatically:

from ace import BrowserUse, LangChain, ClaudeCode

# Browser automation
browser = BrowserUse.from_model(browser_llm=my_llm, ace_model="gpt-4o-mini")

# LangChain chain/agent
chain = LangChain.from_model(my_runnable, ace_model="gpt-4o-mini")

# Claude Code CLI
coder = ClaudeCode.from_model(working_dir="./project", ace_model="gpt-4o-mini")

For direct Anthropic API usage without a runner, compose the SDK steps directly:

from ace import Pipeline, Reflector, SkillManager, Skillbook, learning_tail
from ace.integrations import ClaudeSDKExecuteStep, ClaudeSDKToTrace

skillbook = Skillbook()
pipe = Pipeline([
    ClaudeSDKExecuteStep(model="claude-sonnet-4-20250514"),
    ClaudeSDKToTrace(),
    *learning_tail(Reflector("gpt-4o-mini"), SkillManager("gpt-4o-mini"), skillbook),
])

Shared Features

All runners share these capabilities:

  • Skillbook persistence — save() / load via skillbook_path
  • Checkpointing — automatic saves during long runs
  • Deduplication — prevent duplicate skills
  • Background learning — wait=False for async learning
  • Progress tracking — learning_stats property

Which Integration Should I Use?

  • Building a Q&A or reasoning agent? Use ACELiteLLM
  • Have an existing LangChain chain or agent? Use LangChain
  • Automating browser tasks? Use BrowserUse
  • Running coding tasks with Claude Code? Use ClaudeCode
  • Calling Anthropic directly from your own pipeline? Use Claude SDK
  • Want to monitor costs and traces? Add Opik
  • Learning from OpenClaw session transcripts? Use OpenClaw
  • Exposing ACE as an MCP tool provider? Use the MCP Server and the MCP Client Setup guide
  • Want to send traces to Kayba from your code? Use Tracing
  • Want to use the hosted API instead of running locally? Use the Hosted API CLI
  • Using a different framework? See the Integration Guide to build a custom runner