Skip to content

rangeking_self-evolving-agent.md

Latest commit

 

History

History
42 lines (31 loc) · 2.06 KB

rangeking_self-evolving-agent.md

File metadata and controls

42 lines (31 loc) · 2.06 KB
Raw
repo RangeKing/self-evolving-agent
url https://github.com/RangeKing/self-evolving-agent
content_timestamp 2026-05-26
time_slice 2026-05
timestamp_source web_observed_github_page
collected_at 2026-05-26 05:44:07 +0800
source github

GitHub - RangeKing/self-evolving-agent: An OpenClaw skill that upgrades self-improving agents from reactive error logging to goal-driven capability evolution with curriculum, evaluation, transfer, and promotion.

Source: https://github.com/RangeKing/self-evolving-agent

Raw public page capture notes

This is a raw-style public GitHub page capture recorded by the hourly public metadata update. Shell DNS for api.github.com failed in this sandbox, so the freshness level is web-observed rather than GitHub API verified.

Visible repository metadata

  • Repository: RangeKing/self-evolving-agent
  • URL: https://github.com/RangeKing/self-evolving-agent
  • Stars: 9
  • Forks: 0
  • Commits / release signal: 18 commits, no releases published
  • License: MIT
  • Primary language / stack signal: Python 93.1%, Shell 5.5%, TypeScript 1.4%
  • Collection timestamp: 2026-05-26T05:44:07+08:00

Visible README / page excerpts

  • About text says it is an OpenClaw skill that upgrades self-improving agents from reactive error logging to goal-driven capability evolution with curriculum, evaluation, transfer, and promotion.
  • Visible folders include agents, benchmarks, evals, hooks/openclaw, modules, scripts, and system.
  • README includes a migration path from older .learnings/ material into .evolution/legacy-self-improving/, keeping prior experience as read-only history.
  • Installation examples include cloning into ~/.openclaw/skills/self-evo-agent and bootstrapping an .evolution workspace.
  • Public topics include agent, skill, ai-agent, self-improving-agent, and openclaw.

Raw capture boundary

No benchmark was run, no source clone was modified, and no private or authenticated metadata was used. This file preserves public page evidence for downstream classification, model-card analysis, public reports, and the site index.