Find the right plugin
1,563 verified DSH plugins found
ILRI7512
dsh-autolearn
DSH autonomous-learning plugin (v2): TypeScript engine, 11 native DSH tools, evidence-based verification, optional MCP adapter.
online111111
dsh-hindsight-bridge
Native Hindsight memory for DeepSeek Harness: automatic recall, per-turn retention, safe configuration.
menotbobbybrown
dsh-plugin-memory
Persistent Knowledge Graph & Long-Term Memory Plugin for DeepSeek Harness (dsh) — Everything is a Plugin
MkaliezZ
dsh-repo-map
Lightweight read-only repository structure and symbol map for DSH; complements dsh-context-pack.
SYMlp
dsh-markdown-memory
Markdown folder long-term memory plugin for DeepSeek Harness (dsh): one file per fact, human-editable, git-versionable
futongxu9-maker
dsh-shared-memory
跨对话记忆系统：Hermes 式共享记忆（MEMORY.md + USER.md 注入每个会话系统提示词），memory 工具 + 可视化记忆面板
metallica360qq-web
dsh-work-memory
Cross-conversation work memory for DeepSeek Harness: two-pass topic induction (12 broad + fine labels with editable annotations), segment-level classification, hierarchical recall, and seq-level replay of compacted spans.
mounta1n-18
dsh-context-usage
DeepSeek Harness 插件：/context 斜杠命令 + 上下文用量面板（占用构成、计费用量、轮数/步数/输出速度、账户余额）。零依赖、无构建。 | Zero-dependency DeepSeek Harness plugin: /context command and an at-a-glance context usage panel.
beidou6972yh
dsh-bw-router
B&W (小棒 & 灰狼) — DeepSeek Harness plugin: plan a task into stages, pick the right model per stage, and switch models in-session without losing context.
stevenzhao07
medxpert-reg-hub
Medical-device global regulatory compliance skill hub (Agent Skills): NMPA/FDA/MDR/PMDA pathways & classification, UDI/STED/clinical, CLI + MCP + DSH plugin. 医械全球注册合规技能聚合包（20 枢纽+48 实例实测+CLI+MCP+DSH 插件）
TBAOT
dsh-persona
Multi-persona memory for DeepSeek Harness: two-tier persona memory (user profile + agent notes) with per-turn review. 多个角色各自独立的长期记忆。
zy-xyz
dsh-lcm
Lossless Context Management compaction engine for DeepSeek Harness — LLM leaf summaries over an append-only log, with lcm_expand/lcm_search recovery
swluo777
dsh-fish-dandan
DSH 持久人格与记忆插件：让所有会话保持同一个角色，带三层记忆、心情状态和可选的抖音日常。A DeepSeek Harness plugin for persistent persona, layered memory, mood and optional Douyin automation.
Zhi-yin666
dsh-book
三本书，一副记性 —— 给 AI 的笔记本 / 工作本 / 日记本，全是 Markdown，零依赖
likaileave
dsh-direct-open
Open DSH delivered files, changed-file rows, and file mentions in the system default app, with configurable right-click takeover and a settings page.
ousiqwq
dsh-soul-desk
DeepSeek Harness 人格与记忆管理：多个人设、会话独立切换、跨会话记忆与预设导入。 | Persona and memory management for DeepSeek Harness: multiple personas, per-session switching, persistent memories, and preset imports.
ddlwwc
dsh-rag
DeepSeek Harness 知识库插件：BlockNote 富文本 + markdown 真相源 + 块级笔记 + 角色会话
Lng-ux
dsh-nebula-mind
dsh-nebula-mind — cross-session memory for DeepSeek Harness, and a persistent mind rather than a store. It remembers across restarts and compaction, and forgets the way the sea does: nothing is deleted, things just sink. Hybrid retrieval (BM25 + dense + PPR) fused by RRF, affect as a multiplier, and a galaxy where every edge explains itself.
zhiwuli0228
dsh-memory-hub
Local-first, file-native long-term memory for DeepSeek Harness (DSH). npm: dsh-memory-hub
LBM-OASIG
dsh-grok-bot
Workspace Bot plugin for DeepSeek Harness: resident Bot agents with own sessions, modes, long-term memory, self-waking routines, cross-session messages and multi-Bot rooms.
Psychic-Z
dsh-memory-assistant
记忆助理：长期记忆使用助理。解决记忆越多调用就越费TOKEN的问题，改善AI处理任务时因为上下文压缩造成的记忆缺失。附加好处：节省TOKEN用量
lgquan
dsh-workspace-memory
Durable workspace-scoped memory for DeepSeek Harness and dsh-voco voice agents.
YansongG-HKU
dsh-design-studio
设计台（dsh-design-studio）: Claude Design-class design studio for DeepSeek Harness — 6-step start wizard with intake questions, design-system context, 3 graded HTML variants, Figma-class editor, inline comments, Tweaks, headless-browser checks (0-100 A-F), 5 export formats. Pure Node, MIT.
ousiqwq
dsh-capability-desk
DeepSeek Harness Skill 与 MCP 管理：全局/工作区规则、能力发现、当前模型翻译与缓存。 | Skill and MCP management for DeepSeek Harness: global/workspace rules, discovery, current-model translations, and caching.