Give AI agents new capabilities — without losing control.

Verified with 222+ models

Native runtime for OpenAI, Anthropic, and Gemini. Framework adapters for LangChain, CrewAI, and MCP. Tested against 246 models from 40 providers — 90% achieve perfect tool-calling compatibility.

Native Runtime Integration

Framework Adapters

Compatible (standard Python)

From runtime failure to working capability

Wrap your logic once. AgentNode handles detection, installation, and retry.

Or use detect_and_install() for fine-grained control over detection and installation separately.

Or let the LLM decide

Connect any OpenAI or Anthropic agent to AgentNode. The Runtime registers tools, injects the system prompt, and runs the tool loop automatically.

from openai import OpenAI
from agentnode_sdk import AgentNodeRuntime

runtime = AgentNodeRuntime()
client = OpenAI()

result = runtime.run(
    provider="openai",
    client=client,
    model="gpt-4o",
    messages=[{
        "role": "user",
        "content": "Extract text from report.pdf"
    }],
)
print(result.content)

from anthropic import Anthropic
from agentnode_sdk import AgentNodeRuntime

runtime = AgentNodeRuntime()
client = Anthropic()

result = runtime.run(
    provider="anthropic",
    client=client,
    model="claude-sonnet-4-6",
    messages=[{
        "role": "user",
        "content": "Search for PDF tools"
    }],
)

The LLM discovers, installs, and runs tools autonomously — no hardcoded tool calls. Runtime docs

Not just a registry. A policy-controlled runtime.

Everything your agent needs to grow — under your policies and trust tiers.

Pre-built AI agents that orchestrate tools from the registry to accomplish complex tasks autonomously.