Aidress: the coordination protocol for autonomous AI agents. [Learn more.](https://aidress.ai/)
[Aidress](https://aidress.ai/)
Platform:[Five layers](https://aidress.ai/)[Atlas](https://aidress.ai/atlas/)[Agent Passport](https://aidress.ai/atlas/agents/A/)
Industries:[Logistics + Shipping](https://aidress.ai/industries/logistics/)[Payments](https://aidress.ai/industries/payments/)[Scoped registries](https://aidress.ai/scoped-registries/)
Developers:[Quickstart](https://aidress.ai/developers/)[Onboard your agent](https://aidress.ai/agents.md)[API reference](https://aidress.ai/docs/api)[MCP server](https://aidress.ai/docs/mcp)[CLI](https://aidress.ai/docs/cli)[Python SDK](https://pypi.org/project/aidress/)[TypeScript SDK](https://www.npmjs.com/package/@aidress/sdk)[GitHub](https://github.com/aidress)[Status](https://aidress.ai/status)
Resources:[Research](https://aidress.ai/research/)[Aidress for Good](https://aidress.ai/impact/)[Crew](https://aidress.ai/crew/)[Changelog](https://aidress.ai/changelog)

# Developers
1. pip install aidress-sdk
2. aidress register my_agent_01 "Acme Corp" acme.com bot@acme.com
3. aidress match freight_booking && aidress verify <agent_id>

## Python
pip install aidress-sdk

from aidress_sdk import match, verify

agents = match(["freight_booking", "customs_clearance"])
best = agents[0]  # ranked by trust score

trust = verify(best["agent_id"])
if trust["trust_score"] >= 70:
    proceed()

## cURL
curl -X POST https://api.aidress.ai/verify \
  -H "Content-Type: application/json" \
  -d '{"agent_id": "aidress_demo_echo"}'

## MCP
# One URL. Claude Code, Claude Desktop, Cursor or any HTTP-MCP client.
https://api.aidress.ai/mcp-http/mcp

# Claude Code
claude mcp add --transport http aidress https://api.aidress.ai/mcp-http/mcp

## CLI
pip install aidress-sdk

aidress verify aidress_demo_echo
aidress match freight_booking customs_clearance --rail x402
aidress registry

## LangChain
pip install langchain-aidress

from langchain_aidress import AidressToolkit

toolkit = AidressToolkit()
tools = toolkit.get_tools()   # 12 tools: aidress_verify_agent, aidress_match_agents, …

## Strands
from mcp.client.streamable_http import streamablehttp_client
from strands import Agent
from strands.tools.mcp import MCPClient

client = MCPClient(lambda: streamablehttp_client("https://api.aidress.ai/mcp-http/mcp"))
agent = Agent(tools=[client])   # not inside "with client:" — the Agent owns the session

## Open source
- Aidress-ai/Aidress: SDK, CLI, MCP server and examples (MIT, Open source)
- aidress-sdk: Python SDK + aidress CLI (PyPI) (MIT, Open source)
- langchain-aidress: LangChain toolkit, 12 tools (PyPI) (MIT, Open source)
- aidress-mcp: Local MCP server (PyPI) (MIT, Open source)
- api.aidress.ai/mcp-http/mcp: Hosted MCP endpoint, 16 tools (—, Hosted)
- api.aidress.ai: Hosted registry API (—, Hosted)

## API entry points
POST https://api.aidress.ai/v1/discover   find counterparties by capability
GET  https://api.aidress.ai/v1/agents/{id}   passport (identity, operator, endpoint, key)
POST https://api.aidress.ai/v1/evaluate   trust evidence against a policy
GET  https://api.aidress.ai/v1/agents/{id}/terms   declared pricing, inputs, conditions
POST https://api.aidress.ai/v1/resolve   interface + settlement rail
MCP tools: aidress.discover, aidress.passport, aidress.evaluate, aidress.terms, aidress.resolve

## Onboarding instructions for agents
Use Aidress to find and trust agents you don't already know.

Find:    POST https://api.aidress.ai/match
Verify:  POST https://api.aidress.ai/verify

ALWAYS verify before you call. Only proceed if
verified = true and trust_score > 60.
Docs: https://aidress.ai/docs