Find the right plugin
1,563 verified DSH plugins found
wangzhanchao883
dsh-disk-manager
DeepSeek Harness plugin: scan the whole C drive, classify every big item (safe cache / relocatable / junction-movable / abandoned / config-memory redline / personal), delete/move/redirect on demand with dry-run & undo. C?????????:????+??+?/?/??,????????
wqty123
dsh-bot
Resident agent entity for DeepSeek Harness: bots that live outside any session, with file-based memory you can open and edit, a task queue, an approval gate, and an optional background executor. Ships no platform integrations.好多bug啊，让我再修修吧。感兴趣的可以star一下
liuyuhao1122
dsh-hermes-memory
Lightweight layered memory plugin for DeepSeek Harness with automatic distillation and compaction.
tianyiming1
dsh-plugin-local-prompt-bridge
Stock-safe DSH community plugin: local tokenize pressure + overflow to CONTEXT_WINDOW_EXCEEDED compact-retry
lcthe
dsh-hermes-memory
DSH-native persistent memory and safe session-aware retrieval plugin
jackyytche
dsh-hindsight-memory
Hindsight long-term memory for DeepSeek Harness
yunxiyang
dsh-stepwise-distill
Solidify DSH session history step by step: rewrite tool results in place to the facts the model actually kept, so long conversations stop re-sending noise every turn. Costs tokens and time: a ~770-character contract in every request, plus one extra full-context request and round-trip per step when stepSummary is on.
YanKaFei
Lacan-Knowledge-OS
Corpus-grounded research environment for Lacanian psychoanalysis: frozen scholarly core (39 hash-pinned components), evidence-carrying answers with provenance, MCP surface (10 tools), Obsidian bridge. Engine ships no source text — the reference corpus is a separate public repository, research use only.
having5548
dsh-backup
Backup & restore plugin for DeepSeek Harness: workspace, full conversations (byte-exact), attachments, settings and dsh-mnemon memory data in one zip
RealAlexandreAI
dsh-dejavu-memory
dsh memory: Nocturne Memory client for DeepSeek Harness
JohnXu22786
memory-standard
Memory Standard Protocol (mm) for DeepSeek Harness: layered MEMORY.md, deterministic budgets, frozen snapshots, cross-agent markdown/URI interop; dsh bundle + standalone CLI.
xinshang777
dsh-per-turn-isolation
DSH 插件：让每轮对话互相独立——每轮只把「本轮输入 + 长期设定」交给模型，历史自动折叠，显著降低上下文占用与费用；界面历史仍完整可回看。
JackZo400
dsh-memory-search
Markdown 笔记变成 Agent 的长期记忆：本地 embedding + SQLite FTS5 混合召回，可选密级过滤（dsh 插件）· Long-term memory for dsh agents: hybrid search (local embeddings + SQLite FTS5) over your markdown notes.
Leafstory
dsh-context-checkpoint
为解决 DeepSeek V4.1 Flash 在 DSH 长上下文下丢失结论 / 幻觉升高而做的自动总结-压缩-注入插件：达限 → 总结落盘 → 自动压缩 → 压缩后注入回上下文开头（含 skill 与一键安装器）
liaowenqi123
dsh-meeting-coordinator
DSH 多 Agent「会议室」插件：会议室会籍 + fork 上下文入场 + 主持人控场多轮讨论 + 会后各自压缩纪要并回到工作区继续干活。基于 dsh-std 元协议（Community v0.15）。
YIRC99
dsh-jev-context
面向 DeepSeek Harness 的 JEV 上下文管理插件：按相关度裁剪与召回历史片段，可视化追踪每轮决策和 token 变化。
dreammaker2016
dsh-file-menu
This plugin has no description yet.
CroissanTTs
dsh-bailian-models
[百炼（DashScope）模型目录预置插件]Alibaba Bailian (DashScope) model catalog preset for DeepSeek Harness — 36 models with per-family reasoning effort adaptation, context windows, and auto-adaptation of existing Bailian routes.
hardes11
dsh-squeeze-command
Manual budget-targeted context compression for DeepSeek Harness: the conversation model picks ranges, a cheap flash-tier route writes the summaries.
Dayi-Z
dsh-learn-wiki
边做边学知识库：CRAG 式纠错检索（L1 Markdown wiki + L2 Hindsight）。agent 反复撞同一堵墙时后台限流联网补料，蒸馏落 staged，两段式 commit 才进召回。DSH 插件。
afterrealism
dsh-session-ctxmenu
DSH plugin: right-click a session/project row in the sidebar to open its Rename/Fork/Archive menu
jackeyunjie
dsh-rule-lens
规则透镜（RuleScope Lens）— DSH 本地插件：AGENTS.md/rules 加载可视化、预算治理、硬拦截、Lint、遵守率 | Rule visibility, budget governance & hard guards for DeepSeek Harness
Aik358
dsh-memory-fitting
Intent fitting for DSH: the agent predicts a few directions, asks in batches, and converges one — before it acts. 记忆拟合：先立靶子再问，收敛后提案待确认。
woodfood111
dsh-token-live
Composer readouts for the DeepSeek Harness: CJK-weighted draft token count, session spend at official peak/idle rates, context occupancy, and what the same usage would have cost on the costliest other model; client-only, with no host half and no credential access.