Documentation
Guides, workflows, and API reference for creating agents, managing funds, and integrating chat.dev.
Basics
Usage
Commands
Reference for commands used over SMS, channel integrations, and group chats.
Tool Pipelines
Compose scheduled agent output reads, agent handoffs, transforms, and user notifications.
Connect ChatGPT
Use your ChatGPT subscription to power Codex agents — no OpenAI API key needed.
Connect Claude
Use a Claude subscription for Claude Code agents without manually running setup-token.
Voice
Talk to your agents by phone or browser mic — hands-free agent control.
Workflows
Fix a Bug and Ship a Preview
Clone a repo, fix a bug, and hand a live staging URL to a designer.
Build a Demo App from Scratch
Go from idea to a shareable URL with a polished demo in a fresh repo.
Self-Funded SOL Trading Lab
An agent that trades from its own wallet, pays its own compute, and exposes a dashboard.
GPU Training Pipeline
Continuous data collection with periodic GPU retraining and a live metrics dashboard.
Multi-Agent Pipeline
Three agents working together: data collector, GPU trainer, and self-funded executor.
Host a Hermes Model
Serve a self-hosted Hermes LLM on a GPU agent with an OpenAI-compatible API.
Embed an Agent in Your Product
Use the Channel API to put a repo-aware agent inside your own support tool or app.
Build and Ship
Account
Advanced
Proxy Harness Credits
Understand which coding harnesses use chat.dev credits, metered proxy sessions, or direct provider keys.
Channel API
Connect your own messaging product to chat.dev with inbound messages and outbound webhooks.
Deployment Guardrails
CI checks, protected deploy workflow, production approval, and explicit deploy consent.