dsh-evolve
跨会话记忆，零 token 确定性召回（bigram-Jaccard 与 FTS5 BM25 经 RRF 融合，中文分词精确）；分级审批门只自动确认可逆且源自用户原话的写入，其余待人工确认；重复观察会强化已有理解而非堆叠重复。过程性知识结晶为 SKILL.md，就地精炼升级，并走 active-stale-archived 生命周期治理，归档可逆、可回滚，绝不删除。另含 skill 与记忆双层防臃肿收敛、每轮结束的隔离式后台复盘、以及完整的网页设置页。需要 Node 22.5+。
Cross-session memory with zero-token deterministic recall (bigram-Jaccard fused with FTS5 BM25 via RRF), a tiered approval gate that auto-confirms only reversible user-anchored writes, and reinforcement that strengthens repeated observations. Crystallizes procedures into SKILL.md files that refine in place and are curated through an active-stale-archived lifecycle with reversible archive and rollback. Adds anti-bloat convergence for both skills and memory, background per-turn review in an isolated call, and a web settings page. Requires Node 22.5+.
怎么安装
dsh plugin add github:chenzheshushi-commits/dsh-evolve 简介
dsh-evolve Self-evolving memory and skill lifecycle for DeepSeek Harness. Your agent forgets everything between sessions. This plugin gives it durable memory, turns repeated procedures into reusable skills, and — crucially — keeps that knowledge **from growing into a noise pile**. Real evolution is mutation *plus* selection *plus* pruning; most memory plugins only do the first. The plugin ships **blank**. It has no preloaded opinions about you or your work: only mechanisms and rules. Everything it learns is local to your install and never leaves it. --- Requirements | Requirement | Why | |---|…
推荐参考
信息
- 协议
- MIT
- 语言
- JavaScript
- GitHub 星标
- 4
- 月下载
- –
- 最近更新
- 2026-08-24
- 创建于
- 2026-08-23
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
- ai-agent, deepseek-harness, dsh, dsh-plugin, long-term-memory, memory, self-evolving, skills
同类推荐
hindsight
vectorize-io/hindsight/tree/main/hindsight-integrations/coding-agents
Hindsight：会学习的 Agent 长期记忆系统，自动召回/保存、知识页、深度反思与按仓库隔离的记忆银行。
memsearch
zilliztech/memsearch/tree/main/plugins/dsh
供 DSH 与其他编程 Agent 共享的 Markdown 记忆，支持自动捕获、步骤前上下文注入、搜索召回，以及通过审阅面板实现 memory-to-skill 自进化。