skills.rest
study-yang
study-yangCOMMUNITY
Explore all skills by this author →
GitHub

cloudbase-agent-python

Build and deploy AI agent backends with FastAPI, AG-UI streaming, and LangGraph adapters.

Updated May 14, 2026
Search Tags:#deployment#fastapi#ai agent#langgraph#cloudbase#sse streaming#ag-ui protocol
One-click install
npx skills add https://github.com/study-yang/Vieb-Coding-- --skill cloudbase-agent-python-study-yang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cloudbase-agent-python
Source: https://github.com/study-yang/Vieb-Coding--/tree/main/%E5%BE%AE%E4%BF%A1%E5%B0%8F%E7%A8%8B%E5%BA%8F-%E5%BF%AB%E8%AE%B0/rules/cloudbase-agent/py
Command: npx skills add https://github.com/study-yang/Vieb-Coding-- --skill cloudbase-agent-python-study-yang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires cloudbase-agent-server, cloudbase-agent-langgraph, cloudbase-agent-crewai, cloudbase-agent-coze, langgraph, langchain-openai, fastapi, uvicorn, python-dotenv, pyjwt, and includes references (resource) components.

What problem does it solve? Building a production AI agent backend requires wiring together streaming protocols, framework adapters, authentication, memory, and cloud deployment — this Skill provides the complete CloudBase Agent Python SDK workflow so you don't assemble it from scratch. ## Core Features & Use Cases - Multi-Framework Agent Adapters: Wrap LangGraph, CrewAI, Coze, or custom agent logic behind the AG-UI protocol with SSE streaming and OpenAI-compatible endpoints. - Server & Middleware Patterns: Deploy agents via AgentServiceApp with generator-based middleware for JWT authentication, logging, and user context injection. - Blocking 4-Step Deployment Pipeline: Enforces Python 3.10 setup, atomic env/ dependency builds, integrity verification, and manageAgent deployment to CloudBase. - Use Case: You want to ship a LangGraph chatbot with conversation memory and tool calling as an HTTP service on Tencent CloudBase — this Skill walks you from agent code through verified deployment. ## Quick Start Ask the AI to create a LangGraph agent server with the CloudBase Agent Python SDK and deploy it to CloudBase following the 4-step deployment pipeline.

Frequently Asked Questions about cloudbase-agent-python

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy a LangGraph agent as an HTTP service with Python?▼

Wrap your CompiledStateGraph with LangGraphAgent, then serve it via AgentServiceApp which exposes /send-message (AG-UI SSE) and /chat/completions (OpenAI-compatible) endpoints. Use a MemorySaver checkpointer for conversation persistence.

How to deploy a Python agent to Tencent CloudBase?▼

Use the manageAgent MCP tool with runtime Python3.10 after a blocking 4-step pipeline: ensure Python 3.10, build env/ with a single pip install targeting manylinux2014_x86_64, verify package integrity, then deploy. Do not use createFunction or manageCloudRun.

What is the difference between LangGraph, CrewAI, and Coze adapters?▼

LangGraph adapter suits stateful graph workflows with checkpointing, CrewAI adapter handles multi-agent collaboration, and Coze adapter proxies hosted Coze bots through AG-UI. All implement the same AbstractAgent interface and expose identical endpoints.

Does CloudBase Agent Python SDK support JWT authentication?▼

Yes, via generator-based middleware that decodes the JWT from the Authorization header and injects the user ID into state.__request_context__.user.id. Adapters like CozeAgentAdapter read this reserved field automatically for per-user context.

Why does my deployed agent fail with ModuleNotFoundError?▼

This usually means env/ was built with the wrong Python version or via multiple incremental pip installs, which breaks namespace packages. Fix by deleting env/ and rebuilding with one atomic pip install using Python 3.10.

When should I use CloudRun instead of HTTP Cloud Functions for agents?▼

Use CloudRun only when you need a custom Docker image with system-level dependencies like FFmpeg, resources exceeding Cloud Function limits, or persistent local file storage. Otherwise HTTP Cloud Functions via manageAgent is the default.

Related Skills

Explore More Software Engineering Skills→
PaldomPaldom
1

instrumenting-with-mlflow-tracing

Instrument Python and TypeScript code with MLflow Tracing for LLM calls and agent decisions.

Community
Intermediate
TencentCloudBaseTencentCloudBase
1.1k

cloudbase-agent

Deploy TypeScript AI agents as AG-UI protocol HTTP services on CloudBase.

Official
Advanced
charlesmsiegelcharlesmsiegel

langfuse-strands

Configure Langfuse OTLP exporter to trace AWS Strands Agents.

Community
Intermediate
dang0705dang0705

cloudbase-agent

Automate AI agent development and deployment with the CloudBase Agent SDK.

Community
Advanced