Guide

You've installed Codeg and run your first session. This guide is everything after that — how to get real work done, from the coding surface you live in to the agents, channels, automations, and specialized workflows built on top. (New to the project? The Introduction covers what Codeg is and the thinking behind it.)

The pages are ordered core-outward: start with the Essentials, bring in agents and let them collaborate, then reach for whatever the job needs. Each page stands on its own, so feel free to jump straight to what you're after.

Not sure where to begin?

Read The Workspace to learn the surface every session runs on, then Working with Agents to put an agent to work in it. Curious about the headline feature? Jump to Multi-Agent Collaboration.

Essentials

The surface every Codeg session runs on — learn these first.

• The Workspace — the integrated engineering loop: file tree, editor and diff, git changes, commit, and an embedded terminal, all alongside the agent.
• Conversation Aggregation — pull your existing sessions from every supported agent into one searchable workspace, and pick any of them up where you left off.
• Git & Worktrees — review diffs, stage and commit, manage remotes, and run work in parallel with built-in git worktrees.

Agents

Codeg is agent-agnostic — one consistent interface over many coding CLIs.

• Working with Agents — enable an agent, run its preflight check, and start a session.
• Supported Agents — the full roster, and where each agent keeps its sessions on disk.
• Authentication & Models — sign in with an agent's own plan, point it at a custom endpoint, or use a provider API key, and choose your model.
• Multi-Agent Collaboration — let the main agent delegate to sub-agents of other types within a single task, each running as its own session.

Channels & Automation

Drive Codeg without sitting in front of it.

• Chat Channels — connect Telegram, Lark (Feishu), and iLink (Weixin) to create tasks, approve permissions, and get live updates — all from your chat app.
• Automations — save a fully-configured composer as a reusable automation and run it headlessly, on a cron schedule or on demand.

Extending Codeg

Give every agent more tools and more know-how.

• MCP Servers — add Model Context Protocol servers from a local scan or the built-in registry to expand what your agents can do.
• Skills — install reusable skill packs into a shared store and enable them for the agents you choose, at global or project scope.

Domain Workflows

Purpose-built modes for specific kinds of work.

• Office Documents — create, analyze, proofread, and edit .docx, .xlsx, and .pptx files through the bundled officecli, with live in-tab preview.
• Scientific Research — a curated set of research skills, from hypothesis and experimental design to analysis and write-up, that any agent can invoke.
• Project Boot — scaffold a new project visually, configuring it on the left while a live preview updates on the right.

Running Codeg on a server instead of the desktop? Everything here works the same in the browser — see Deployment and Configuration.