
useagent
Hand off the work. Get back the result. The open-source AI coworker for your team: agents with their own cloud computer, your tools and context, handing back…
Install with your AI
Paste into Claude Code, Cursor, or any agent — it reads the repo and wires the tool into your project.
Install and set up useagent (git-clone project) into my current project. Found on https://claudeers.com/useagent Repo: https://github.com/useagenthq/useagent Homepage/docs: https://useagent.org Detected install method: git-clone → git clone https://github.com/useagenthq/useagent Category: devtools. Platforms: cli, api, desktop, web. Read the repo's README for exact setup and env vars, then install it and wire it into my project. Claudeers Health Verdict: active; community-verified: false. Confirm the source before running anything.
git clone https://github.com/useagenthq/useagent
// compatibility
| Platforms | cli, api, desktop, web |
|---|---|
| Operating systems | — |
| AI compatibility | claude |
| License | AGPL-3.0 |
| Pricing | open-source |
| Language | TypeScript |
useAgent
Hand off the work. Get back the result.
The open-source AI coworker for your team.
Quick Start · Self-hosting · Documentation · Architecture
Alpha software. useAgent is under active development: expect rough edges, and APIs/schemas may change between releases. It already runs real daily workloads, but pin a tag if you need stability.
useAgent turns the AI agents you already use - Claude Code, Codex, OpenCode - into coworkers with their own cloud computer, your tools, and your company context. They hand back finished work, not just answers: live websites, decks, spreadsheets, research reports, and tested pull requests.
Under the hood: durable, threaded sessions in isolated Linux sandboxes. One event contract, one UI, your infrastructure.
Every run is an event-sourced timeline in Postgres: it survives restarts, renders through one session grammar regardless of engine, and stays inspectable after the fact.
Features
- Any agent, one interface - Codex (API key or ChatGPT subscription), Claude Code, and OpenCode behind one provider-neutral event contract.
- Durable runs - runs survive backend restarts; recovery re-probes live sessions and adopts finished work instead of failing it.
- Real sandboxes - each thread gets an isolated Linux workstation: terminal, repositories, browser, and a visible desktop (noVNC) with MP4 recording. Daytona and Cube behind one sandbox contract.
- Trusted gateway - credentials never enter the sandbox; knowledge, memory, skills, GitHub, web search, desktop control, and artifact publishing are typed tools served by the backend.
- Human-in-the-loop - destructive tools pause on an approval card (web or Slack) and resume with a one-shot, argument-bound capability.
- Skills from GitHub - import
SKILL.mdfiles as versioned skills, auto-resync, and rank them into every turn. - Knowledge, wiki, and team memory - org-scoped retrieval with citations and a human-reviewed learning lane.
- Slack-native - mention-to-run, threaded replies, attachments, artifacts, and approvals in the thread.
- Native artifacts - DOCX, XLSX, PPTX, and PDF with revisioned editing and native renderers.
Integrations
Work arrives from anywhere and tools stay behind the gateway:
| Channels in | Web app, Slack, REST API, schedules - every channel enters through the same run door |
| Native | Slack (mentions, threads, approvals, delivery), GitHub (App auth, clones, PRs) |
| Via connectors | Gmail, Linear, Notion, HubSpot - OAuth handled by the broker, tokens sealed server-side |
| Workspace surfaces | Knowledge base, team memory, skills and playbooks, scheduled automations |
Quick Start
Requires bun and Postgres 16+ with the pgvector extension (stock Postgres images do not include it). No Postgres handy? One container does it:
docker run -d --name useagent-pg -p 5432:5432 -e POSTGRES_HOST_AUTH_METHOD=trust pgvector/pgvector:pg16
export DATABASE_URL=postgres://postgres@localhost:5432/postgres
for workspace in \
packages/agent-harness packages/artifact-workspace \
packages/agent-client packages/artifact-formats packages/conformance \
backend frontend; do
(cd "$workspace" && bun install)
done
bun run dev:backend # API + orchestration on :3201
bun run dev:frontend # UI on :3400 (proxies /api/* to the backend)
bun run typecheck covers every package.
Self-hosting
useAgent runs on any Linux host - AWS, Google Cloud, Azure, Hetzner, or
bare metal. See infra/self-host/ for the full
guide, including the one-command Hetzner reference host (Terraform) and the
provider-agnostic deploy-app.sh:
SERVER_IP=<host-ip> PG_PASSWORD=... OPENROUTER_API_KEY=... \
infra/self-host/deploy-app.sh /path/to/this/repo
Sandboxes are pluggable: Daytona (managed service - pairs with a host on
any cloud, easiest start) or CubeSandbox (self-hosted runtime on your own
hardware - full data locality). Production deploy lanes are documented in
infra/self-host/; the provisioning is provider-agnostic
and addresses the host over SSH.
Architecture
Three properties do the heavy lifting:
- The engine is a plug. Claude Code, Codex, OpenCode, and Pi all speak one canonical event contract through the engine adapters - swap engines and your threads, artifacts, and memory stay.
- Every run is an event log. Postgres is the source of truth: runs survive backend restarts, replay exactly, and stay inspectable after the fact.
- Credentials never enter the sandbox. The agent's computer is isolated; every integration call crosses the trusted gateway as a typed tool, and the keys live only on your control plane.
| Path | What it owns |
|---|---|
frontend/ | Product UI: chat, sessions, skills, playbooks, wiki, artifacts, automations, settings |
backend/ | Control plane: auth, runs, sandboxes, engines, knowledge, memory, artifacts, connectors |
packages/ | Shared contracts: thread events, canonical engine events, workpieces, renderers |
docs-site/ | Documentation site: concepts, architecture, API, operations |
infra/self-host/ | Self-hosting on any provider + Hetzner reference Terraform |
memory/ | Optional team-memory service |
Deeper reading: the documentation site and the interactive request-flow diagram.
The additive immutable-container release lane is documented in
docs/operations/immutable-releases.md.
License
useAgent is free and open-source software under the GNU AGPL v3.0 (AGPL-3.0-only). You may use, modify, and self-host useAgent under the AGPL.
If you want to embed useAgent into proprietary software, distribute it without AGPL obligations, build an OEM or white-label product, or obtain different terms, a commercial license is available.
Contributions are accepted under the CLA. The useAgent name and logo are covered by the trademark policy, not the code license. Third-party components are listed in NOTICE; vendored and ported files carry per-file attribution headers.
// faq
What is useagent?
Hand off the work. Get back the result. The open-source AI coworker for your team: agents with their own cloud computer, your tools and context, handing back finished work - websites, decks, spreadsheets, reports, PRs. Runs Claude Code, Codex, OpenCode on your subscription.. It is open-source on GitHub.
Is useagent free to use?
useagent is open-source under the AGPL-3.0 license, so it is free to use.
What category does useagent belong to?
useagent is listed under devtools in the Claudeers registry of Claude-compatible tools.
// embed badge
[](https://claudeers.com/useagent)
// retro hit counter
[](https://claudeers.com/useagent)
// reviews
// guestbook
// related in Developer Tools
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Curs…
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA
AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs,…