From 7425cabd321f3dc2b52b4bbd1a90822d511ba89e Mon Sep 17 00:00:00 2001 From: chendanyang Date: Sun, 6 Sep 2026 08:31:42 +0800 Subject: [PATCH] Add AgentDescent to evolution AgentDescent is a parallel, asynchronous optimizer for self-evolving agents: N workers propose diffs to a shared, version-controlled library of skills, prompts and harness modules, and an aggregator resolves conflicts, fuses complementary diffs and accepts merges on a Beta posterior over held-out reward. MIT, 203 stars, active, zero required dependencies (Python >= 3.9). Co-Authored-By: Claude Opus 5 --- README.md | 1 + data/projects.json | 13 +++++++++++++ 2 files changed, 14 insertions(+) diff --git a/README.md b/README.md index dbe7b69..7207d84 100644 --- a/README.md +++ b/README.md @@ -58,6 +58,7 @@ Projects focused on enabling AI agents to evolve, learn, and improve autonomousl - [**OpenProgram**](https://github.com/Fzkuji/OpenProgram#readme) - Self-programming AI agent framework whose agents create, run, and refine their own workflows while the runtime manages models, tools, memory, context, and multi-agent collaboration. by [@Fzkuji](https://github.com/Fzkuji) (350 stars) - [**Reef**](https://github.com/Human-Agent-Society/reef#readme) - Continual learning infra for self-improving agents. Serves agent traffic, turns matched feedback into model-weight or harness updates, and publishes accepted updates as versioned artifacts. by [@Human-Agent-Society](https://github.com/Human-Agent-Society) (343 stars) - [**SEAgent**](https://github.com/SunzeY/SEAgent#readme) - Self-Evolving Computer Use Agent with Autonomous Learning from Experience. by [@SunzeY](https://github.com/SunzeY) (263 stars) +- [**AgentDescent**](https://github.com/Birfy/agentdescent#readme) - Parallel, asynchronous optimizer for self-evolving agents: workers propose diffs to a shared library of skills, prompts and harnesses, and an aggregator resolves conflicts and accepts merges on a Beta posterior over held-out reward. by [@Birfy](https://github.com/Birfy) (203 stars) ## Memory Systems diff --git a/data/projects.json b/data/projects.json index 4a464c9..ba2edf6 100644 --- a/data/projects.json +++ b/data/projects.json @@ -1524,6 +1524,19 @@ ], "stars": 343 }, + { + "name": "AgentDescent", + "repo": "Birfy/agentdescent", + "description": "Parallel, asynchronous optimizer for self-evolving agents: workers propose diffs to a shared library of skills, prompts and harnesses, and an aggregator resolves conflicts and accepts merges on a Beta posterior over held-out reward.", + "category": "evolution", + "maintainer": "Birfy", + "tags": [ + "self-evolving", + "self-improving", + "research" + ], + "stars": 203 + }, { "name": "Autohand Code", "repo": "autohandai/code-cli",