convex-agents

Build persistent AI agents in Convex with threads, tools, and streaming.

1|Updated Oct 10, 2025
One-click install
npx skills add https://github.com/cgRGM/rivercitymd --skill convex-agents-cgrgm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: convex-agents
Source: https://github.com/cgRGM/rivercitymd/tree/main/.cursor/skills/convex-agents
Command: npx skills add https://github.com/cgRGM/rivercitymd --skill convex-agents-cgrgm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you build persistent AI agents in Convex without stitching together ad hoc chat storage, tool calls, or retrieval logic. It focuses on durable conversational state, real-time responses, and organized orchestration so agent features remain reliable as the app grows.

Core Features & Use Cases

  • Persistent threads and message history for chat experiences that survive refreshes and restarts.
  • Streaming assistant responses for responsive user interfaces.
  • Tool integration and workflow orchestration for actions like searching knowledge, creating tasks, or triggering business logic.
  • RAG patterns for assistants that need to search stored documents and generate grounded answers.
  • Use it to add a support bot, research assistant, or internal ops assistant to a Convex app.

Quick Start

Use the convex-agents skill to design a Convex-backed agent with persistent threads, streaming replies, and tool-based workflows for your application.

Frequently Asked Questions about convex-agents

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

FAQPage Schema
How do I build stateful AI agents in Convex with persistent chat history?

Stateful AI agents in Convex are built using persistent threads and message history, ensuring conversational state survives app refreshes and restarts for reliable chat experiences.

Can I stream AI chat responses to a user interface using Convex?

Yes, streaming assistant responses in Convex provides real-time updates to user interfaces, creating responsive chat experiences without waiting for full message generation.

How do I integrate tool execution and workflow orchestration into Convex AI agents?

Tool integration and workflow orchestration in Convex agents enable actions like searching knowledge, creating tasks, or triggering business logic through structured tool calls.

What is the best way to implement RAG retrieval for an AI assistant in Convex?

RAG retrieval in Convex uses vector search to query stored documents, allowing research assistants to search knowledge bases and generate grounded answers for user queries.

Does Convex support durable state for production AI customer support bots?

Convex supports durable state for production customer support bots, managing conversational threads and tool execution reliably as application scale and complexity grow.

When should I not use Convex for AI agent chat orchestration?

Convex agent orchestration requires agent component setup, Convex functions, and vector search configuration, so simpler stateless chat apps may not need this durable state infrastructure.