#307307th of 324 in Skills, plugins and prompts
evo
Plugin that runs parallel optimization experiments on a codebase via coding agents
- Stars
- 1.5k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Jul 2026
- Language
- Python
2 on 10 Apr 20261,490 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
evo installs as a plugin for coding-agent hosts such as Claude Code, Codex and Cursor. Its discover skill finds a benchmark and gates, then optimize runs a tree search of experiments, each in its own git worktree, keeping changes that improve the score. A local dashboard (default port 8080) tracks experiments, and runs can go to Modal, E2B, Daytona, AWS, Azure or SSH.
Who it is for: Engineers who want agent-driven benchmark optimization of an existing codebase
Strengths
- Gates discard experiments that beat the score but fail tests or a held-out floor
- Tree search with five frontier strategies, including argmax, top_k and pareto_per_task
- Supports eight agent hosts and eight execution backends, local and remote
- Dashboard for monitoring experiments, frontier strategy and backend selection
Weaknesses
- Sends anonymous telemetry by default; opt out with evo telemetry off
- CLI, skills and hooks must stay version-matched; drift caused silent failures earlier
- Needs a separate host agent CLI and its model access; cost is not documented
- Remote backends require extra installs and provider accounts; setup steps are not covered
What it needs
- no GPU
- Needs uv, git, Claude Code, Codex, Cursor, Kimi, OpenClaw, Hermes, Opencode or Pi, Modal, E2B, Daytona, AWS, Azure or SSH (optional, remote runs)
- port 8080
Quick answers
Is evo open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Also in Skills, plugins and prompts
See all 324| Rank | Project | Score |
|---|---|---|
| 1 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 96 out of 100 |
| 2 | skillsAgent skills for planning, TDD, debugging and code review | 95 out of 100 |
| 3 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 95 out of 100 |
| #4 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 95 out of 100 |
| #5 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 95 out of 100 |