Delegate a code task
Describe the goal; the task runs in a managed cloud environment with build, test, terminal, and preview workflows, and produces a reviewable change.
How execution worksOPEN SOURCE · BUILT FOR ENGINEERING TEAMS
AI coding, from task to review.
Start coding with DeepSeek V4.1 Flash, already built in. Describe a task, run it in a cloud workspace, and review the change — free to start.
FREE TO STARTFree cloud dev environment + 10M tokens/day (resets daily) — no install, run in the browser. Plan terms are set by the vendor; verify current limits.
The MonkeyCode task workspace
Text-edited English mock-up of the original product screenshot, not a real English runtime capture. Interface and account values vary by version.
What teams run on it
Build, organize, and keep work moving.
Delegate a code task, work with documents, or follow progress across devices. Explore the workflow that fits your team.
Tame docs and engineering research
With the MonkeyWork desktop agent, authorized local files (Word, PDF, Markdown) can be summarized, rewritten, and organized; a paired browser extension assists gathering web material.
Desktop mode boundariesFollow up from any device
Start a task on the web, continue it on the desktop, and check progress, keep chatting, or receive results on mobile while the work keeps running.
See the feature mapExecution modes
Pick where the work runs.
Use a managed cloud environment, work with local files on desktop, or operate the platform on your own infrastructure.
Hosted web
Tasks run in managed cloud dev environments. Browser only, no install; free tier to start.
Cloud env · per-plan quotasDesktop client (MonkeyWork)
A local agent works on authorized local files and can continue cloud tasks; browser pairing assists web work.
Windows · macOS · LinuxSelf-hosted
Run the platform inside your network with centralized admin control. Verify model egress — it is a deployment condition, not a data guarantee.
Online/offline installer · AGPL-3.0Full boundary table with sources: Execution-mode boundaries →
REPRODUCIBLE EXERCISE · V1
Your first task: fix, test, explain
Start with a tiny Python project. You can inspect the input, make one bounded change and reproduce the checks before connecting a real repository.
From the official service
From developers using MonkeyCode.
Quotes are published feedback from the official hosted service; some details are generalized for privacy. monkeycode-ai.net/ ↗
What I value most is that it is device-agnostic: a computer at the office, a tablet at home, a phone to check progress. Tasks keep running.
Setting up environments used to take a lot of time and felt tedious. After using MonkeyCode a lot of that wasted effort disappeared—it works out of the box and lets me focus on the actual product.
Product notes · checked 2026-09-08Read the update log →
News and research
Guides for your next engineering decision.
Source-checked AI coding news plus durable guides on agents, governance, and self-hosting.
MonkeyCode desktop: a bounded local-directory workflow
Use one disposable example folder for desktop tasks, inspect file changes independently and distinguish local execution from model-provider data processing.
Connect GitHub to MonkeyCode, then verify revocation
A source-reviewed GitHub App workflow covering a single repository, project creation, missing repositories and separate installation and user authorization revocation.
MonkeyCode on Ubuntu: preflight, installation and acceptance
A concrete Ubuntu workflow with a tested preflight script, a versioned installer observation, expected checks and failure recovery for a self-hosted evaluation.
Direct answers
The short version.
Can I use DeepSeek V4.1 Flash for free?
Yes. MonkeyCode includes DeepSeek V4.1 Flash with free access, subject to your account’s current usage limits. Direct use of the DeepSeek API is a separate service with its own billing.
What is MonkeyCode?
An open-source AI development platform for engineering teams: managed agent tasks, server-side development environments, model choice, team visibility, and a private-deployment path.
Where do tasks actually run?
Hosted tasks run in managed cloud environments. The desktop client adds a local agent for authorized directories. Self-hosted installs run on your infrastructure — with model egress to verify.
Can MonkeyCode be self-hosted?
Yes. Published starting minimums are 2 CPU cores / 4 GB / 40 GB for the console plus 8 CPU cores / 16 GB / 100 GB for a development environment host.
Is self-hosting enough to keep all code private?
Not automatically. It keeps the platform and task records inside your network, but model endpoints, telemetry, credentials, and backups must be verified per configuration.
What does the free tier include?
The published Basic plan is $0 with a cloud dev environment and 10M tokens per day that reset daily. Verify current terms — promotional quotas can end.
START BUILDING