convex-agents

Develop persistent, stateful AI agents with Convex for long-running conversations and tool-driven automation.

Updated Apr 12, 2026

Search Tags:#ai #llm #agents #tools #workflows #convex

Ready to run

npx skills add https://github.com/althof3/TCG-auction --skill convex-agents-althof3

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: convex-agents
Source: https://github.com/althof3/TCG-auction/tree/main/.agent/skills/convex-agents
Command: npx skills add https://github.com/althof3/TCG-auction --skill convex-agents-althof3

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Building AI agents with Convex often requires stitching together persistent state, thread management, tool execution, streaming responses, RAG, and durable workflows into a cohesive, scalable solution. This Skill provides a guided approach to assemble these capabilities into a single, reusable agent framework.

Core Features & Use Cases

Persistent stateful agents with thread management to maintain context across sessions.
Real-time streaming responses to provide a responsive UX for conversations.
Tool integration to execute Convex functions and external actions within agent workflows.
Durable workflows for long-running tasks with reliable progress tracking.
Retrieval-augmented generation (RAG) with vector search for knowledge retrieval.

Quick Start

Install the convex agent package and initialize a basic agent example to start a chat workflow.

Frequently Asked Questions about convex-agents

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

FAQPage Schema

How do I build persistent AI agents with Convex?▼

You build persistent AI agents with Convex by using a modular framework to stitch together stateful thread management, streaming responses, tool execution, and durable workflows for long-running conversations.

What is the best way to maintain context across long-running AI conversations?▼

The best way to maintain context across long-running AI conversations is using persistent stateful agents with thread management. This approach tracks session state reliably to preserve conversational history.

Can I use RAG and vector search for knowledge retrieval in Convex agents?▼

Yes, you can use retrieval-augmented generation (RAG) with vector search for knowledge retrieval in Convex agents. This integration allows the agent to query external knowledge bases during workflows.

Does Convex support real-time streaming responses for AI chat applications?▼

Yes, Convex supports real-time streaming responses for AI chat applications. This feature provides a responsive user experience by delivering conversational outputs incrementally as they generate.

How do I execute external actions and functions within durable AI workflows?▼

You execute external actions and functions within durable AI workflows through safe tool integration. This allows agents to trigger Convex functions and external services with reliable progress tracking.

Do I need a modular architecture to scale multi-thread AI agent collaboration?▼

Yes, a modular architecture is required to scale multi-thread AI agent collaboration. It enforces safe, reusable tool integration and streaming responses, ensuring cohesive and scalable agent operations.