dsh-skill-evolution
观察 agent 执行轨迹，在每个成功回合结束时按信号触发 LLM 评审，把值得复用的工作流结晶为已注册技能，并随后续运行持续改进。
Watches agent execution traces and fires an LLM review at every successful turn end when signals trip, crystallizing reusable workflows into registered skills that improve progressively with later runs.
怎么安装
dsh plugin add github:VanadisGithub/dsh-skill-evolution 简介
dsh-skill-evolution English | 中文 A **skill self-evolution plugin** for DeepSeek Harness (DSH): it watches agent execution traces and, at the end of every successful turn, fires an LLM review when signals trip — distilling workflows worth reusing into **crystallized skills** registered in the skill catalog. Later runs of the same workflow fold fresh lessons into the existing skill — skills are alive and get better with use. Highlights **Event-driven crystallization** — review fires at turn end, not on pure frequency counting; any of three signals (complex / recovered / repeated) sends the turn …
推荐参考
信息
- 协议
- MIT
- 语言
- JavaScript
- GitHub 星标
- 0
- 月下载
- –
- 最近更新
- 2026-09-02
- 创建于
- 2026-09-01
基础安全检查
- 检查结果
- curated 收录但无 npm 包/安装命令
- 收录来源
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- 主题标签
- agent-memory, cordis, deepseek-harness, dsh, dsh-plugin, plugin, self-evolution, skill
同类推荐
mcp-server-chart
antvis/mcp-server-chart
🤖 一个可视化MCP和技能，包含使用@antvis的25+可视化图表。用于图表生成和数据分析。
avoid-ai-writing
conorbronsdon/avoid-ai-writing
审计并重写内容以去除AI写作模式的技能。可与你最喜欢的代理一起使用，包括Claude Code、OpenClaw、Codex和Hermes。
ruflo
ruvnet/claude-flow
🌊 原始的代理元框架。部署智能多智能体群体，协调自主工作流，构建对话式AI系统。具备自适应记忆、自学习智能、RAG集成，并原生集成Claude Code / Codex / Hermes等。