back to the wire
ANative-Lab

EvoAgentX

🚀 EvoAgentX: Building a Self-Evolving Ecosystem of AI Agents

Written by ANative-Lab, whose words those are. Agentwire did not build this and has not tested it.

open the repository

WHAT VOUCHED FOR IT

  • StarredPast the absolute star floor.
  • AgenticCarries two or more agentic topics in its own repo metadata.

WHAT THE HEALTH CHECK READ

Read on Sep 29, 2026, today. Every line links to the request the value came from. Nothing below was run, and the score is explained on the criteria page.

WHAT THE README DECLARES

This is one tool’s stated scope. Several permitted tools running in one agent compose into outcomes none of them permits alone, and nothing here measures that. What the permission field is not

HOW THE SCORE WAS MADE

  • Last commit0.91share 0.22
  • Commits in 90 days0.20share 0.20
  • Last release0.99share 0.18
  • Issue close time0.00share 0.15
  • Licence0.00share 0.10
  • Install command1.00share 0.15

The score is 0.57 out of 1, the weighted mean of the factors that were measurable, which came to 1.00 of a possible 1.00 in weight. That puts it in active.

WHERE THE ACCOUNTS POSTED IT

BlueskySep 28, 2026the post

EvoAgentX builds multi-agent workflows from a prompt, scores them with task-specific evaluators, and iteratively optimizes them. ANative-Lab's framework is aimed at researchers and workflow engineers. #SoftwareArchitecture