nowfound

AI · October 12, 2026

I made an agent can self-improving while working

An open-source framework that evaluates and refines your AI agent’s skills, tools, and workflows against real-world tasks. Get your work done at top GEAR.

I made an agent can self-improving while working launched on October 12, 2026 with 5 votes, #722 of 3,296 launches that month and more than 75% of that year's launches.

Visit yehuitang0316.github.io ↗Alternativestop 22% of October 2026

What it does

Don’t let your work wait. GEAR brings recursive self-improvement (RSI) to your agent’s harness — using task feedback to continually refine skills, tools, and workflows for the work that matters. On these two real tasks, Luna max with a GEAR-tuned harness completes requirements that Astra max with the general-purpose Codex agent misses. Watch the difference side by side. An optimized harness can change what an agent gets done. Explore the results reported in our open experiments. Task pass rate also increased from 27% to 53% . Astra max with Codex scored 84.08% objective completion and 57% task pass rate. Luna’s reported task pass rate increased from 52.87% to 84.26% . These runs compare…from yehuitang0316.github.io

Does a similar job

all alternatives →
  • SolidSep 2026 · solid.tech · ▲471

    Agents with their own computers, accounts, and budgets.

  • skills.shJan 2026 · skills.sh · ▲318

    The Open Agent Skills Ecosystem

  • AgentspanMay 2026 · orkes.io · ▲93

    Open-source runtime for durable AI agents

  • AI Agent Skills RefinerMar 2026 · skills-refiner.com · ▲85

    Skills with 210k GitHub Data & Translate/Refine &Benchmark

  • Output.ai - OSS framework we extracted from 500+ production AI agentsApr 2026 · output.ai · ▲40

    Open-source TypeScript framework for AI dev: prompts, evals, tracing, cost tracking, security, and durable execution. Built for Claude Code.

  • Open-source implementation of Stanford's self-learning agent framework2025 · github.com · ▲10

    We implemented Stanford's Agentic Context Engineering paper which shows agents can improve their performance just by evolving their own context. How it works: Agents execute tasks, reflect on what worked/failed, and curate a "playbook" of strategies. All from execution feedback - no training data needed. Happy to answer questions about the implementation or the research!

More ai this month

the category →
  • Video model that turns script into viral social media videos

    AI · 17d ago · spira.ai

  • The browser where Claude Code and Codex work beside you

    AI · 17d ago · rill.love

  • Kolibri is Aleph Alpha

    AI · 19d ago · tej.as

  • Paintings by AI models: programs run through a simulation of oil paint.

    AI · 20d ago · stillwet.art

  • A real-time world model you can explore and change

    AI · 29d ago · world.pixverse.video

  • Hemory▲381

    Keep listening. Searchable memory for your AI agents.

    AI · 30d ago · hemory.com

Launched alongside, October 2026

the whole month →