DSH-PLUGIN STORE / LIVE CATALOG
聚合 GitHub 上的 DSH 插件，打造 DeepSeek Harness 生态的一站式目录。
32 个项目，匹配「token」
MemTensor
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
foryourhealth111-pixel
Intelligent Skill routing and workflow orchestration for AI agents — +21.12 pp reward, −29.6% tokens on SkillsBench with DeepSeekV4Flash-VE.
toolclub
Persistent multi-model workflow teams for DeepSeek Harness — dynamic lead planning, bounded DAGs, per-agent model/tools, Run Center and Token insights.
WestFox-AwA
DSH 插件 · 注入式优化器 0.8（主线）：你照常说话，它在你发送后,AI接收前把"这一轮到底要什么"理清楚，再把这份理解交给工作 AI(上下文注入) —— 原话不改写，条条带逐字依据。含控制界面（档位/权限/模型/上下文/只读工具）、拦截浮层（思维层+产出层）与真实 token 用量。可明显提升大多数模型的发挥稳定性,尤其DeepSeek-V4.1-Flash,可有明显提升
Zoria-Lind
Layered token-optimization pipeline for DeepSeek Harness: output ladder, MCP lazy loading,compaction driver, cache-hit reporting. Built on real DSH plugin APIs; ~40-60% input saved in long sessions.
phoenixlucky
让 Agent 高效又守纪律 — 不止省 token：ZeroToken 压缩无效上下文/推理/输出；尉缭子十原则约束权限边界、单一指令、先谋后动、验证先于结束；附 Unicode 编码规范、搜索规范、六种任务模式。More than token savings: ZeroToken efficiency + AI coding discipline for Reasonix / Codex / OpenCode / Hermes
1420079678-ctrl
Gates 84.7% of tool-schema prompt tokens away for DeepSeek Harness: plugins declare themselves organs, a nerve impulse routes each command, reflex arcs fire with zero model calls, memory consolidates while idle, every failure is attributed before retry. Reproducible offline, MIT.
chenzheshushi-commits
Self-evolving memory + skill lifecycle for DeepSeek Harness — durable cross-session memory with zero-token deterministic recall, tiered approval, reinforcement learning from repetition, and anti-bloat convergence for both skills and memory.
ZekaiShi
Unified DeepSeek Harness plugin: role-based subagent routing + per-agent evolution (prefercmd/memory as knowledge allow/deny lists), so repeated tasks start from proven commands and save tokens. Unified subagent routing and evolution: prefercmd/memory serve as knowledge allow/deny lists, saving tokens.
lhh010
DSH Web UI 会话进度插件：输入框停靠区常驻进度条（todos 真实进度 / token 速率与 ETA / 中断橘红态 / 后台任务与子代理青色态 / Token 用量徽标与缓存命中面板），窄窗口两行自适应，零核心改动
yindf
Context folding for DeepSeek Harness (DSH): wrap work in named tasks, fold finished spans into short summaries, recall originals on demand.
Lee-Hilex
基于 MinerU 的 DeepSeek Harness 多模态文档解析插件：PDF/Word/PPT/Excel/HTML/图片 → 结构化 Markdown（免 Token 的 Agent 轻量解析 / 精准解析双 API 模式）
PianoPrince
拖拽真迁移 DSH 会话跨工作区：批量移动、分组合并、挂错归位、回收站、备份恢复 + 对话内 agent 工具；步步备份可回滚，零 token / True migration of DSH sessions across workspaces: bulk move, group merge, misfiled homing, recycle bin & backup restore + conversational agent tools; every step backed up, zero tokens
Context Assembler DSH — DeepSeek Harness 上下文汇编插件：按话题块编排上下文，最大化每 token 互信息密度
skepsun
极简·省 token 的 DSH 长期记忆插件：零 LLM 自动捕获、符号索引渐进披露、ESR 证据闭环。Minimalist token-saving memory for DeepSeek Harness: zero-LLM auto-capture, symbolic index, ESR-lite evidence closure.
Grivn
Long-term memory for AI agents on Jev. Keep raw records, judge them with a fast System 1 model and answer from under 4k tokens of context.
ItBayMax
Replay a DeepSeek Harness session at its original token cadence — an in-app playback theater with play/pause/step/speed/seek.
Sev7eEn7
DeepSeek Harness（DSH）上下文管理与 token 优化插件：工具输出过滤、历史上下文裁剪、技能按需披露；离线回放请求载荷减少 30% 以上（按字符数）。
Snow-ea
Deterministic, recoverable tool-result compression and cache-aware compaction for DeepSeek Harness.
TuringCorp-net
Generic stateless dialogue compression that mimics human memory. LLM conversations stay bounded forever — no session management, no context overflow. Ships a ready-to-use adapter for DeepSeek Harness (DSH).
helloxkk
Context visibility + GitHub-style usage heatmap for DeepSeek Harness (dsh) — Codex-style /context lens + 53-week contribution graph
Ln1m
DSH composer suite: skill sets — one switch swaps the skills injected into a session · DSH 输入区家族：技能档——切档即换本会话注入的技能清单
deepseekv41flash
给 DSH 模型每轮回填宿主实测的上下文占用，并阻止"上下文快用尽"的错判被压缩摘要继承。
viyiviyi
树形任务流 · DeepSeek Harness 插件：把多步任务组织成目标→任务→子任务三级树，节点完成时由 AI 提交结果，插件随即把该节点的执行过程从上下文里折叠掉。
