Clevrr Computer is live · free · last verified 30 Oct 2026 · viability monitor
Clevrr Computer
Clevrr Computer is an open-source AI desktop agent that automates mouse, keyboard, and screenshots with PyAutoGUI.
Pick Clevrr Computer when you want to read, run, or fork a working screenshot-reason-act loop and you're comfortable cloning a repo and wiring API keys into a .env file. The get_screen_info and PythonREPLAst tools are plain enough to trace the whole decision chain, which is its real value. Pass if you need unattended execution against live business systems — the maintainers themselves flag beta status and prompt-injection override risk.
Verdict generatedVerified 10h ago · liveness 65/100 · cite: rightaichoice.com/tools/clevrr-computer
- Developers exploring AI desktop automation and computer-use agents
- Students learning multi-modal screen-vision reasoning from readable code
- Engineers scoping an in-house computer-use build against a working baseline
- Hobbyists prototyping desktop automation locally on a throwaway VM
- Non-technical users wanting plug-and-play desktop automation
- Production enterprise deployments that need reliability and support guarantees
- Mobile or web-only automation, since it drives a desktop OS
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Clevrr Computer if you need unattended desktop automation against a live business system — the maintainers label it beta and flag prompt-injection override risk in their own README.
Model access is on you: the tool ships no inference, so your Google Gemini or Azure OpenAI usage bills directly to your own account.
Clevrr Computer itself is $0 — MIT-licensed source you clone from GitHub. Your real spend is inference: bring your own Google Gemini API key, or Azure OpenAI credentials for --model openai, and pay that provider's rates. That makes it cheaper than commercial computer-use products with subscription tiers, but also means no support contract, no SLA, and no roadmap. For teams that need someone accountable when the agent clicks wrong, the paid alternatives are the comparison to make.
In short
Clevrr Computer — Clevrr Computer is an open-source AI desktop agent that automates mouse, keyboard, and screenshots with PyAutoGUI. Best for Developers exploring AI desktop automation and computer-use agents, Students learning multi-modal screen-vision reasoning from readable code, Engineers scoping an in-house computer-use build against a working baseline. Free to use.
What is Clevrr Computer?
Clevrr Computer — Clevrr Computer is an open-source AI desktop agent that automates mouse, keyboard, and screenshots with PyAutoGUI. Best for Developers exploring AI desktop automation and computer-use agents, Students learning multi-modal screen-vision reasoning from readable code, Engineers scoping an in-house computer-use build against a working baseline. Free to use.
Clevrr Computer is a small, MIT-licensed GitHub project from the Clevrr-AI organization that turns an LLM into a desktop automation agent. Give it a task and it reasons over screenshots, then acts on your machine — moving the mouse, clicking, typing, firing keyboard shortcuts, capturing the screen, and managing windows through Python's PyAutoGUI library. Everything is driven from a terminal command, `python main.py`, after a clone, `pip install -r requirements.txt`, and dropping API keys into a `.env` file. The architecture is a constant screenshot-reason-act loop. A `get_screen_info` tool grabs the current screen and overlays a coordinate grid so the model can reason about true pixel positions, and a multi-modal LLM answers the agent's questions about what it sees. The agent builds a chain of thought and executes each step through a `PythonREPLAst` tool that runs PyAutoGUI calls. Execution results feed back into the next decision, which is how it copes with a UI that didn't respond the way it expected. Model choice is a flag away. Gemini is the default driver, and `--model openai` switches to Azure OpenAI using four `AZURE_OPENAI_*` environment variables. A floating TKinter overlay sits above your screen showing live progress, disabled with `--float-ui 0`. The repo is modest — 317 stars, 44 forks, 9 commits — so it reads as a working reference implementation rather than a packaged product. The maintainers label it beta and are blunt about the risk: content on webpages or in images can override your instructions via prompt injection. Their own advice is a dedicated VM or container with minimal privileges, no sensitive credentials near it, an allowlist of domains, and a human confirming consequential actions. Positioned next to commercial computer-use agents, Clevrr Computer trades polish and support for readable source code you can fork.
Key features
- Mouse automation: move, click, and drag via PyAutoGUI
- Keyboard automation: type text and fire shortcut inputs
- Screenshot capture with a coordinate-grid overlay for vision input
- Window management: minimize, resize, and close
- Multi-modal screen understanding from captured screenshots
- Chain-of-thought reasoning loop that plans and executes task steps
- get_screen_info tool that queries current on-screen state
- PythonREPLAst tool that executes PyAutoGUI actions as Python
- Error handling with a feedback loop feeding results into the next step
- Gemini model support as the default driver
- Azure OpenAI support via the --model openai flag
- Floating TKinter overlay showing real-time agent progress
- Disable the floating overlay with --float-ui 0
- Configuration through a .env file for API keys and endpoints
- MIT license for cloning, forking, and modification
Behind the verdict
The best way to learn how a computer-use agent actually works is to read one. Clevrr Computer is small enough to finish in an afternoon and honest enough to show the ugly parts — the coordinate grid, the REPL tool that executes PyAutoGUI calls, the feedback loop when an action fails. If you're evaluating the category or scoping an in-house build, this is a legible baseline. We'd reach for it first when the deliverable is understanding, not a deployed bot. In practice it runs from a terminal: clone, install requirements, rename `.env_dev` to `.env`, add a `GOOGLE_API_KEY` or the Azure OpenAI values, then `python main.py`. The floating TKinter overlay keeps progress on screen; pass `--float-ui 0` when it gets in the way. Switching drivers is a single flag, so you can compare how Gemini and Azure OpenAI handle the same task without touching code. Where it bites is reliability. This drives a real desktop, and a misread screenshot means a click in the wrong place. The maintainers call it beta and warn that instructions on webpages or inside images can override your prompt — prompt injection is a live concern, not a footnote. Their mitigation list is sensible and you should follow it: isolated VM or container, minimal privileges, no saved logins, allowlisted domains, human sign-off on anything with real-world consequences. Who should pass? Anyone who wants plug-and-play desktop automation, a supported product with SLAs, or something to point at financial transactions and credential-bearing accounts. There is no mobile or web-only story here either, since it automates a desktop OS. Non-technical users will bounce off the setup. Compared with commercial computer-use offerings, Clevrr Computer doesn't compete on guardrails or managed infrastructure — it competes on
Models under the hood
as of 2026-10-24
Limitations
- Clevrr Computer is beta, and the maintainers say so in the README's caution block.
- It drives the real cursor and keyboard through PyAutoGUI rather than an application API, so a misread screen can produce a misplaced click.
- The README warns that it poses risks distinct from standard APIs or chat interfaces, that these are heightened when it interacts with the internet, and that instructions on webpages or contained in images may override your instructions or cause mistakes — the prompt-injection failure mode, stated as a known behavior.
- Recommended mitigations: a dedicated VM or container with minimal privileges, no sensitive data such as account login information, internet limited to an allowlist of domains, and a human confirming decisions with real-world consequences such as accepting cookies, executing financial transactions, or agreeing to terms of service.
- The repository is early-stage — roughly 318 stars, 44 forks, and 9 commits — with 4 open issues and 3 open pull requests, so third-party support is thin and the README is the documentation.
as of 2026-10-10
Researching Clevrr Computer? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is Clevrr Computer still active?
Clevrr Computer is live · free · last verified 30 Oct 2026 · viability monitor.
Viability score
How well maintained and how widely used is Clevrr Computer? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →What do users say about Clevrr Computer?
What people actually say about Clevrr Computer — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
Researched35 mentions across 3 sources (YouTube, Product Hunt, GitHub) · researched Aug 30, 2026.
Average across the 3 sources that answered — each source counts once, not each post.
- +Open-source and MIT-licensed, allowing full customization and learning.
- +Supports both Gemini and Azure OpenAI, offering model flexibility.
- +Local execution with no cloud dependency, preserving privacy.
- +Real-time floating UI shows agent progress, aiding transparency.
- +Clear command-line arguments for model and UI toggling.
- −GitHub issues report model selection conflicts (Gemini vs OpenAI).
- −Missing Pillow from requirements.txt causes setup failures.
- −Send button sometimes not visible, blocking interaction.
- −Initialization can hang, leaving the agent stuck.
- −Limited community support with 7 open issues and no roadmap.
- • API costs for Gemini or OpenAI usage
- • Time spent troubleshooting setup issues
How we verified this
Every listed fact about Clevrr Computer— pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 30 Oct 2026. Pages mix model-assisted synthesis with human review; see how we verify.
Verification history
We have re-verified Clevrr Computer 11 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed pricing tiers, summary, description, our verdict
- — changed pricing page
- — changed pricing page
- — changed features, who it suits, who should skip it, summary, pricing tiers, description, our verdict
- — changed who should skip it, pricing tiers, our verdict, use cases, summary, models, features, setup time, hidden costs, description, pricing fit, limitations, switching in, switching out, pricing page
- — changed features, pricing tiers, summary, description, who it suits, our verdict, who should skip it
Showing the 6 most recent of 11 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does Clevrr Computer cost in 2026?
Clevrr Computer: Free to use. Pricing last verified 30 Oct 2026.
Plans compared
Every published Clevrr Computer tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Open Source (MIT) | Free | — | — |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where Clevrr Computer's pricing actually pencils out — and where peers do it cheaper.
Clevrr Computer itself is $0 — MIT-licensed source you clone from GitHub. Your real spend is inference: bring your own Google Gemini API key, or Azure OpenAI credentials for --model openai, and pay that provider's rates. That makes it cheaper than commercial computer-use products with subscription tiers, but also means no support contract, no SLA, and no roadmap. For teams that need someone accountable when the agent clicks wrong, the paid alternatives are the comparison to make.
Has Clevrr Computer's pricing changed?
No plan price for Clevrr Computer has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick Clevrr Computer?
Pick Clevrr Computer if you are Developers exploring AI desktop automation and computer-use agents, Students learning multi-modal screen-vision reasoning from readable code, Engineers scoping an in-house computer-use build against a working baseline, Hobbyists prototyping desktop automation locally on a throwaway VM.
Look elsewhere if you are Non-technical users wanting plug-and-play desktop automation, Production enterprise deployments that need reliability and support guarantees, Mobile or web-only automation, since it drives a desktop OS, Workflows touching logins, credentials, or financial transactions.
Skip it if Skip Clevrr Computer if you need unattended desktop automation against a live business system — the maintainers label it beta and flag prompt-injection override risk in their own README.
GeneratedReal-world workflow fit
Concrete scenarios for the personas Clevrr Computer actually fits — and what changes day-one when you adopt it.
Clone the repo into a throwaway VM, rename .env_dev to .env, paste a GOOGLE_API_KEY, run pip install -r requirements.txt, then python main.py and watch the floating TKinter overlay narrate the screenshot-reason-act loop.
Outcome: You see how a multi-modal model turns a coordinate-grid screenshot into a PyAutoGUI call, and you have readable source to modify.
Read main.py and the utils folder to see how get_screen_info, PythonREPLAst, and the error-handling feedback loop are wired, then run --model openai with the four AZURE_OPENAI_* variables against your own deployment.
Outcome: You decide within a session whether to fork this or build from scratch, and you know which driver fits your existing cloud contract.
Run a plain-language multi-step desktop task in a sandboxed VM with the floating overlay visible, then re-run with --float-ui 0 to show the headless path.
Outcome: The audience sees a screenshot go in and a real click come out, with the chain-of-thought steps visible on screen.
Use cases
- Automate repetitive data entry across desktop applications by driving mouse and keyboard input
- Simulate clicks and keystrokes to test GUI interactions inside a sandboxed VM
- Prototype a computer-use agent before committing to a from-scratch build
- Demonstrate AI-driven desktop automation during a class or internal presentation
- Run a multi-step desktop task described in plain language through the chain-of-thought loop
- Study how a screenshot-reason-act loop is wired together in readable Python
Setup time & first value
How long it actually takes to get something useful out of Clevrr Computer — broken out by persona, not the marketing-page minute.
Developers: 15-30 minutes to first action if Python and pip are ready — clone, pip install -r requirements.txt, rename .env_dev to .env, add GOOGLE_API_KEY, python main.py. Add 10-20 minutes if you're switching to --model openai, since that needs all four AZURE_OPENAI_* values. Students and hobbyists without a Python environment or a spare VM: budget an hour or more. Standing up the recommended
Resources & guides
Tutorials & Learning
YouTube returned 6 videos for “Clevrr Computer”, and we withheld 6: 6 did not mention Clevrr Computer. We are showing none, because we could not prove any of them are about Clevrr Computer.
Official links
What are the best Clevrr Computer alternatives?
The closest alternatives to Clevrr Computer we list are NeuralAgent, Agent.exe and Cyberdesk. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| NeuralAgent | Freemium | Pricing not published | — |
| Agent.exe | Free | Free | — |
| Cyberdesk | Contact Sales | Custom pricing | — |
Alternatives to Clevrr Computer
View all
NeuralAgent
NeuralAgent is a computer-use AI agent that runs mouse-and-keyboard workflows inside SAP, legacy desktop apps, and browsers.
Tools that pair well with Clevrr Computer
Common stack mates teams adopt alongside Clevrr Computer, with the specific reason each pairing earns its keep.
NeuralAgent
NeuralAgent is a computer-use AI agent that runs mouse-and-keyboard workflows inside SAP, legacy desktop apps, and browsers.
Agent.exe
Open-source Electron app that lets Claude's computer-use API drive your Mac's mouse, keyboard, and screen.
Cyberdesk
Cyberdesk automates legacy Windows desktop apps — EHRs, ERPs, portals — with plain-English computer-use agent workflows.
Clevrr Computer vs its closest rival
Clevrr Computer's closest head-to-head is with Locus Robotics — Locus Robotics is the clear choice for warehouses needing proven, scalable automation with 2-3x productivity gains and enterprise support, but requires ongoing RaaS investment. Clevrr Computer offers a free, open-source desktop automation agent for developers, but is not suitable for production or non-technical users. Choose based on whether you need physical warehouse automation or AI-driven desktop scripting. Verdict generated
Head-to-head comparisons
Clevrr Computer vs Locus Robotics
Locus Robotics is the clear choice for warehouses needing proven, scalable automation with 2-3x productivity gains and enterprise support, but requires ongoing RaaS investment. Clevrr Computer offers a free, open-source desktop automation agent for developers, but is not suitable for production or non-technical users. Choose based on whether you need physical warehouse automation or AI-driven desktop scripting.
GeneratedClevrr Computer vs Presto Voice
Choose Presto Voice if you run a QSR chain and need a proven drive-thru voice AI with upselling and POS integration. Choose Clevrr Computer if you're a developer wanting to experiment with open-source desktop automation using AI agents. Presto is enterprise-ready; Clevrr is a beta tool for tinkerers.
GeneratedClevrr Computer vs Truleo
Truleo is the clear choice for law enforcement agencies needing a production-ready AI platform to unify siloed data and generate investigative leads. Clevrr Computer is a free, open-source experiment for developers exploring desktop automation, but it's far from enterprise-grade. Most buyers should choose Truleo.
GeneratedFrequently asked questions
Used Clevrr Computer? Help shape our editorial sentiment research.