Skip to content

08-contribution-workflow-and-production-governance.md

Latest commit

 

History

History
194 lines (142 loc) · 6.41 KB

08-contribution-workflow-and-production-governance.md

File metadata and controls

194 lines (142 loc) · 6.41 KB
layout default
title Chapter 8: Contribution Workflow and Production Governance
nav_order 8
parent Qwen-Agent Tutorial

Chapter 8: Contribution Workflow and Production Governance

Welcome to Chapter 8: Contribution Workflow and Production Governance. In this part of Qwen-Agent Tutorial: Tool-Enabled Agent Framework with MCP, RAG, and Multi-Modal Workflows, you will build an intuitive mental model first, then move into concrete implementation details and practical production tradeoffs.

This chapter closes with contribution strategy and team governance patterns.

Learning Goals

  • contribute examples and framework improvements responsibly
  • maintain docs/tests along with behavior changes
  • enforce secure defaults in deployment playbooks
  • preserve observability and auditability in production setups

Governance Checklist

  • review tool and MCP scopes before rollout
  • pin model/config versions per environment
  • maintain run logs and evaluation evidence
  • include security review for sandboxed execution components

Source References

Summary

You now have a complete Qwen-Agent path from first setup to production governance.

Next tutorial: Mini-SWE-Agent Tutorial

Source Code Walkthrough

qwen_server/utils.py

The read_meta_data_by_condition function in qwen_server/utils.py handles a key part of this chapter's functionality:

def read_meta_data_by_condition(meta_file: str, **kwargs):
    if os.path.exists(meta_file):
        with open(meta_file, 'r', encoding='utf-8') as file:
            meta_info = json.load(file)
    else:
        meta_info = {}
        return []

    if 'url' in kwargs:
        if kwargs['url'] in meta_info:
            return meta_info[kwargs['url']]
        else:
            return ''

    records = meta_info.values()

    if 'time_limit' in kwargs:
        filter_records = []
        for x in records:
            if kwargs['time_limit'][0] <= x['time'] <= kwargs['time_limit'][1]:
                filter_records.append(x)
        records = filter_records
    if 'checked' in kwargs:
        filter_records = []
        for x in records:
            if x['checked']:
                filter_records.append(x)
        records = filter_records

    return records

This function is important because it defines how Qwen-Agent Tutorial: Tool-Enabled Agent Framework with MCP, RAG, and Multi-Modal Workflows implements the patterns covered in this chapter.

qwen_server/utils.py

The save_history function in qwen_server/utils.py handles a key part of this chapter's functionality:

def save_history(history, url, history_dir):
    history = history or []
    history_file = os.path.join(history_dir, get_basename_from_url(url) + '.json')
    if not os.path.exists(history_dir):
        os.makedirs(history_dir)
    with open(history_file, 'w', encoding='utf-8') as file:
        json.dump(history, file, indent=4)


def read_history(url, history_dir):
    history_file = os.path.join(history_dir, get_basename_from_url(url) + '.json')
    if os.path.exists(history_file):
        with open(history_file, 'r', encoding='utf-8') as file:
            data = json.load(file)
            if data:
                return data
            else:
                return []
    return []

This function is important because it defines how Qwen-Agent Tutorial: Tool-Enabled Agent Framework with MCP, RAG, and Multi-Modal Workflows implements the patterns covered in this chapter.

qwen_server/utils.py

The read_history function in qwen_server/utils.py handles a key part of this chapter's functionality:

def read_history(url, history_dir):
    history_file = os.path.join(history_dir, get_basename_from_url(url) + '.json')
    if os.path.exists(history_file):
        with open(history_file, 'r', encoding='utf-8') as file:
            data = json.load(file)
            if data:
                return data
            else:
                return []
    return []

This function is important because it defines how Qwen-Agent Tutorial: Tool-Enabled Agent Framework with MCP, RAG, and Multi-Modal Workflows implements the patterns covered in this chapter.

examples/assistant_add_custom_tool.py

The MyImageGen class in examples/assistant_add_custom_tool.py handles a key part of this chapter's functionality:

# Add a custom tool named my_image_gen：
@register_tool('my_image_gen')
class MyImageGen(BaseTool):
    description = 'AI painting (image generation) service, input text description, and return the image URL drawn based on text information.'
    parameters = [{
        'name': 'prompt',
        'type': 'string',
        'description': 'Detailed description of the desired image content, in English',
        'required': True,
    }]

    def call(self, params: str, **kwargs) -> str:
        prompt = json5.loads(params)['prompt']
        prompt = urllib.parse.quote(prompt)
        return json.dumps(
            {'image_url': f'https://image.pollinations.ai/prompt/{prompt}'},
            ensure_ascii=False,
        )


def init_agent_service():
    llm_cfg = {'model': 'qwen-max'}
    system = ("According to the user's request, you first draw a picture and then automatically "
              'run code to download the picture and select an image operation from the given document '
              'to process the image')

    tools = [
        'my_image_gen',
        'code_interpreter',
    ]  # code_interpreter is a built-in tool in Qwen-Agent
    bot = Assistant(
        llm=llm_cfg,

This class is important because it defines how Qwen-Agent Tutorial: Tool-Enabled Agent Framework with MCP, RAG, and Multi-Modal Workflows implements the patterns covered in this chapter.

How These Components Connect

flowchart TD
    A[read_meta_data_by_condition]
    B[save_history]
    C[read_history]
    D[MyImageGen]
    E[init_agent_service]
    A --> B
    B --> C
    C --> D
    D --> E
Loading