English | 中文

ANPCrawler Examples - AMAP Service

This directory contains examples using ANPCrawler to access AMAP agent services.

Contents

• amap_crawler_example.py - Complete example showcasing all ANPCrawler features
• simple_amap_example.py - Simplified example for quick start
• README.md - This documentation file

Prerequisites

Environment Setup

# Option 1: Install via pip
pip install anp

# Option 2: Source installation (recommended for developers)
git clone https://github.com/agent-network-protocol/AgentConnect.git
cd AgentConnect
uv sync

DID Authentication Files

Ensure the following files exist:

• docs/did_public/public-did-doc.json
• docs/did_public/public-private-key.pem

Running Examples

Simple Example

uv run python examples/python/anp_crawler_examples/simple_amap_example.py

Complete Example

uv run python examples/python/anp_crawler_examples/amap_crawler_example.py

Example Features

1. Fetch Agent Description Document

# Access URL and retrieve ad.json content
content_json, interfaces_list = await crawler.fetch_text(
    "https://agent-connect.ai/agents/travel/mcp/agents/amap/ad.json"
)

2. Parse JSON-RPC Interfaces

# Automatically parse interfaces and convert to OpenAI tool format
tools = crawler.list_available_tools()

3. Call JSON-RPC Methods

# Call discovered tools/methods
result = await crawler.execute_tool_call(tool_name, arguments)

Example Output

After running examples, you will see:

1. Agent Description Document Content - Complete ad.json content
2. Discovered Interfaces - JSON-RPC interfaces extracted from agent description
3. Available Tool List - Names of callable tools
4. Tool Call Results - Actual JSON-RPC call return results

Troubleshooting

File Not Found Error

FileNotFoundError: DID document file not found

Solution: Ensure the following files exist:

• docs/did_public/public-did-doc.json
• docs/did_public/public-private-key.pem

Network Connection Error

Ensure your network can access the agent-connect.ai domain.

Authentication Failure

Check if DID document and private key files are correctly generated and matched.

Code Structure

# 1. Initialize crawler
crawler = ANPCrawler(
    did_document_path="path/to/did.json",
    private_key_path="path/to/private-key.pem"
)

# 2. Fetch agent description
content, interfaces = await crawler.fetch_text(url)

# 3. List tools
tools = crawler.list_available_tools()

# 4. Call tools
result = await crawler.execute_tool_call(tool_name, arguments)

Related Documentation

• ANPCrawler API Documentation
• DID WBA Authentication Examples
• Project Root README