Are There Open-Source AI Agents for Code Development? (2026)
Yes. As of Aug 6, 2026, out of 235 AI agents for Code development tools in the KanonAgent index, 57 are judged open-source (every judgement requires a source quote); this page covers the top 10 by real traction.
Yes, 12 open-source AI agents match code development needs. They enable autonomous editing, testing, and workflow automation while keeping all code and models inspectable. Open-source status here means public repositories with no paywalls for core features.
Updated 2026-08-06 · 10 products · live data from KanonAgent
1. gh-aw7 upvotes
gh-aw automates GitHub code reviews, PRs, and branch ops as an open-source workflow agent.
2. goose52.4k upvotes
goose installs, executes, edits, and tests code with any LLM as a fully open-source extensible agent.
3. adk-python21.0k upvotes
adk-python provides an open-source Python toolkit to build and deploy custom coding agents from source.
4. opencode414 upvotes
opencode autonomously writes, tests, and fixes code as an open-source collaborative coding partner.
Agent-run executes coding tasks safely in a sandbox as an open-source agent for automated development.
6. deer-flow79.4k upvotes
deer-flow brings Gemini-powered code generation and debugging directly into the terminal as open-source.
7. qwen-code26.8k upvotes
qwen-code runs in the terminal to write, debug, and refactor code with native open-source support.
8. jcode8.0k upvotes
jcode offers an open-source framework to build and manage multiple coding agents in one workflow.
9. oh-my-pi6.2k upvotes
oh-my-pi delivers terminal-based code edits with subagents and LSP as an open-source agent.
10. gptme4.4k upvotes
gptme runs locally in the terminal to write code and execute commands as a customizable open-source agent.
How to choose
Prioritize agents with active GitHub repos and recent commits for long-term maintenance under open-source. Check LLM compatibility since some require local models while others integrate cloud ones. Avoid near-misses like retrieval-only tools when full autonomous coding is needed. Test sandboxing features first if security matters for your codebase. Verify token usage and hardware demands before scaling to large projects.
What the data says
Computed from our index over the 10 products on this page; judgement fields are left blank where we cannot read them (methodology).
AutonomyL4 × 1 · L2 × 6 (7/10 judged)
Prerequisitesopen source × 10 · self-hostable × 5 · bring your own API key × 2
In the index since2026-07-08 — 2026-07-30
Ranked by real traction from our index — not editorial picks, and no paid placement. Every judgement field requires a source quote; where we cannot read it, we leave it blank. Full criteria, thresholds and known limits: methodology.
FAQ
Which ones run fully locally without API keys?
goose, gptme, and qwen-code support local execution with open models.
How do I contribute or fork these agents?
All listed have public GitHub repos; start with goose or deer-flow for high activity.
Do any integrate directly with VS Code or GitHub?
gh-aw focuses on GitHub workflows while others work via terminal or custom extensions.