Docs/Overview
Everything DGC does, grouped. Start at the top, or jump straight to the part you need.
DGC documentation#
DGC is a local-first coding agent for your terminal and for VS Code. It runs a real agentic loop — planning, reading files, running tools, editing code — against a model you choose: a local endpoint, any OpenAI-compatible host, or a coding subscription you already pay for. Your code and your prompts go only where you point them.
New to DGC?
Install it, launch it, and connect a model — about five minutes.
Getting started#
Getting started
Install, first launch, connect a model.
Keyboard shortcuts
Every key binding in the composer + transcript.
Slash commands
The full / command reference.
Command line
Flags, subcommands, and one-shot runs.
Using DGC#
Permission modes
Default · acceptEdits · plan · auto.
Plan mode
Read-only planning, then one-tap approve.
Sessions & rewind
Resume, jump, and undo whole turns.
Standing goals
Persistent objectives with an explicit lifecycle.
Providers & models#
Connect your model
Point DGC at Ollama, llama.cpp, vLLM, or a cloud host.
Subscriptions
Bring your own Claude / Codex / Qwen / Kimi / Copilot plan.
Thinking & reasoning
Off · low · medium · high · xhigh, and preserving it.
Features#
Artifacts
Preview what the agent builds on a localhost URL.
MCP servers
Connect external tools over MCP.
Lifecycle hooks
Run observable commands at agent boundaries.
Skills
Reusable instruction packages.
Multiple agents
Run a fleet of agents at once + the dashboard.
Training export
Your real sessions → scrubbed fine-tuning JSONL.
Python code-action (power mode)
A persistent Python interpreter for token-efficient work.