dsh-experience-library
Experience validation layer for DSH: zero-token collection of tool-failure/retry/search signals, AI refinement into verified skill books (three-layer verification), looked up at task start. Ships 10 trial skills and benchmark data (complex-task success 100% vs 60% bare).
DSH 经验验证固化层：零 token 采集工具失败/重试/搜索信号，加工轮提炼成经三层检验的技能书，任务开局翻书复用。内置 10 本试验经验 + benchmark 数据（复杂任务成功率 100% vs 裸环境 60%）。
How to install
dsh plugin add dsh-experience-library About
dsh-experience-library — 更有经验的 DeepSeek English | 中文 **meow-memory 让你的 AI 记住,我让你的 AI 记住"对的做法"。** > 本项目的方法论与调度机制深受 dsh-meow-memory(@Phant0Meow) 启发: 全自动空闲派发机制参考了它的 dream 调度(窗口表 + idleMinutes + 租约防重),并作为**适配器**对接它的 lesson 层。 思路起点:记忆插件解决"记住",而本插件解决"记住**对的做法**"——验证过滤幻觉,固化可复用的操作流程。 为什么需要它 meow-memory 解决"**记住**"(陈述性记忆);经验库解决"**记住对的做法**"(程序性经验,验证过滤幻觉) **三层检验**:层1 清单可见 → 层2 触发翻书 → 层3 至少 3 个新会话抽样且成功率 ≥2/3 = "已验证" **双轨判定**:结果导向为主(行为正确 + 一次成功即通过)+ 措辞指纹为辅(技能书签名句,如"先查地图,再下铲子") **Benchmark 验证**:复杂任务上经验库成功率 100% vs 裸环境 60%、耗时快 4.7 倍、thinking 降 77%(详见 Benchmark 章节) 经验分层(重要) | 层级 | 内容 | 说明 | |---|---|---| | **核…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 1
- mo. downloads
- 228
- Last push
- 2026-08-27
- Created
- 2026-08-24
Links
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- dsh-plugin
Related plugins
superpowers
obra/superpowers
An agentic skills framework & software development methodology that works.
dev-browser
SawyerHood/dev-browser
A Claude Skill to give your agent the ability to use a web browser
ruflo
ruvnet/claude-flow
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated