Urgent.News

What's breaking now, across thousands of outlets.

AI

Building a $5.70/Month Autonomous AI Coding Agent in the Cloud Using Open-Source Terminal Agent Management

Originally published on tamiz.pro . Introduction Autonomous AI coding agents promise to reduce developer friction by running continuously in the cloud, writing, testing, and refactoring code without human intervention. Commercial offerings like GitHub Copilot Workspace or Replit Agent cost $10–$20/month per seat. This guide shows how to build an equivalent autonomous coding agent for just…

Autonomous AI coding agents aim to cut developer overhead by running independently in the cloud, performing tasks like coding, testing and refactoring without human oversight. Commercial services such as GitHub Copilot Workspace or Replit Agent charge $10–$20 per month per user. This article demonstrates building a comparable autonomous coding agent for just $5.70 per month, leveraging open-source tools agent-deck and orca, on a basic cloud virtual machine.

The process begins with setting up a lightweight Ubuntu virtual machine on a budget cloud platform. DigitalOcean and Vultr are examples of such providers. A minimal Ubuntu 22.04 droplet with one CPU, one gigabyte of RAM, and 25 GB of storage is provisioned at around $5 per month. After logging into the machine via SSH, the system is updated and basic tools like Git, curl, Python, and Docker are installed.

Next, the agent-deck open-source terminal session manager is installed. This software provides a tmux-like interface and can manage multiple AI agent sessions. The orca CLI runtime, another essential component, is then installed. orca supports plugin-style workflows and integrates seamlessly with agent-deck.

To enable autonomous operations, the agent is configured to push changes to a GitHub or GitLab repository automatically. SSH key authentication is set up for secure access, and the OpenAI API key is securely exported to the environment. The agent is then configured to write code, run tests, and commit improvements based on prompts given by the agent-deck session.

Security measures include restricting shell access to SSH keys only, using a dedicated GitHub deploy key with write-only permissions, and running the agent under a non-root user. Cron jobs are set up to restart the agent daily, ensuring its continuous operation.

In summary, building an autonomous AI coding agent for under $6 per month is achievable with open-source tools agent-deck and orca, significantly undercutting commercial alternatives. This setup empowers developers by automating coding, testing, and code management tasks, potentially increasing productivity and reducing operational costs.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Declare `tools` Only When Narrowing an APX Agent

Declare tools Only When Narrowing an APX Agent An agent definition should describe a durable role without freezing a temporary snapshot of the runtime around it.

  • Omitting "tools" field preserves broad default allowlist from runtime.
  • Declared "tools" list acts as narrowed allowlist for constrained agents.
  • Remove "tools" field if it's just available tools at agent creation.

How to Run an AI Model on Almost Any Hardware: A Practical Guide to Tiny Local LLMs

This month a developer ran a working language model on a Sony PSP-2000. That is a 333 MHz MIPS handheld from 2007 with 64 MB of RAM, and it is streaming English text onto its LCD at one to two tokens…

  • Developer ran language model on 2007 Sony PSP handheld device with 333 MHz MIPS processor
  • Karpathy's stories15M model int8-quantized to 17 MB, runs in 1,100 lines of C code
  • Guide provides hardware-first approach for running tiny local LLMs on various hardware

More from Sunday 6 September →