BigHugger
GH Repository · evo-hq

evo

turns your codebase into an autoresearch loop — discovers what to measure, instruments the benchmark, then runs tree search with parallel subagents.

stars
1,459
30-day movement
+72/day
Related entries
60
Connections
1
skill-packPythonllm-agentsevolutionary-algorithmscodexcode-optimizationclaude-codeautoresearchautonomous-agentsagent-skills

evo is a skill-pack that turns a codebase into an autoresearch loop: it discovers what to measure, instruments a benchmark, then runs tree search with parallel subagents. It is a Python repository under evo-hq, licensed Apache-2.0.

Use it when you want autonomous, evolutionary search over code improvements driven by benchmarks rather than manual tuning.

Use it to

  • Instrument a benchmark for your codebase
  • Run tree search with parallel subagents
  • Apply evolutionary algorithms to code optimization
  • Use as a skill pack with Claude Code or Codex

For Developers using LLM agents for code optimization

Role
skill-pack
Language
Python
Licence
Apache-2.0
Forks
110
Open issues
16
Last push
2026-07-17
Latest release
v0.3.0 · 2026-04-26
topicsautoresearchcode-optimizationevolutionary-algorithmsllm-agentsagent-skillsclaude-code