DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Community Feature Requests — Plugin Hub News
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in

COMMUNITY CHALLENGES

Find your next challenge.

Real problems from the community. Find one that sparks an idea, explore the discussion, and build something people need.

Showing 41–60 of 255 community requestsUpdated hourly from community discussions
All updatesDSH ReleasesFeature Requests
All3 days7 days14 days
TopNew
OpenShippedSolved by pluginWon't fixAll statuses
#Problems worth solvingCommunity demand
41
PLUGIN
41

Configure Per-Request Timeouts

Support configurable longer timeouts for large-payload RPCs and model requests, including slow local model loading, generation, retries, and large outputs, while retaining shorter defaults for ordinary calls.

Explore challenge ↗

The problem

Support configurable longer timeouts for large-payload RPCs and model requests, including slow local model loading, generation, retries, and large outputs, while retaining shorter defaults for ordinary calls.

Start with these discussions

  • llm-pi-ai Timeout increase for local LLMsgithub · ▲17 · 21 comments
  • 使用本地部署的模型时，如果重启过电脑，Vram内容被清空，则发起对话时模型要思考很久时间，如果超过五分钟就报错重试了，导致无法继续对话github · ▲1 · 4 comments
  • dsh-client-connection：history 等大载荷 RPC 受全局 30s 超时限制，Web 端报「历史加载失败：signal timed out (internal)」—— timeoutPolicy 钩子已存在但无人接线github · ▲1 · 1 comments
Last active Sep 19, 2026
4 authors5 posts · 20 reactions27 comments
42
HARNESS
2

Refresh MCP tools after reconnect

Allow long-lived sessions to re-fetch and register MCP tools after a connection failure or successful reconnect, so recovered servers can be used without restarting the session.

Explore challenge ↗

The problem

Allow long-lived sessions to re-fetch and register MCP tools after a connection failure or successful reconnect, so recovered servers can be used without restarting the session.

Possibly addressed by:Live Mcp · Mcp Skill Control · Mcp Skill Control

Start with these discussions

  • mcp-client: stateful MCP server restart leaves every tool call failing until DSH restartsgithub · ▲2 · 4 comments
  • Headless session env: are DSH_-prefixed variables intentionally dropped before reaching the session?github · ▲1 · 1 comments
Last active Oct 15, 2026
2 authors2 posts · 3 reactions5 comments
43
PLUGIN
6

Configure SSRF-Safe Fake-IP Resolution Allowlisting

Allow web-fetch plugins to configure trusted resolvers, domains, or CIDR ranges for transparent-proxy fake-IP DNS while retaining SSRF protection. Support relevant IPv4 and IPv6 pools, keep private and link-local networks blocked by default, and provide actionable errors.

Explore challenge ↗

The problem

Allow web-fetch plugins to configure trusted resolvers, domains, or CIDR ranges for transparent-proxy fake-IP DNS while retaining SSRF protection. Support relevant IPv4 and IPv6 pools, keep private and link-local networks blocked by default, and provide actionable errors.

Possibly addressed by:Web Fetch Enhanced · Fakeip Compat · Fetch Third Party

Start with these discussions

  • web-fetch-http: add a config toggle to allow non-public DNS answers (fake-ip proxy compatibility)github · ▲12 · 4 comments
  • 建议为 web_fetch 增加可配置的非公网 CIDR 白名单，以支持 Fake-IP 与特殊路由环境github · ▲6 · 1 comments
  • [Bug] web_fetch 在 fake-ip（198.18.0.0/15）环境下对所有站点都抛 WEB_BLOCKED_URL，且无配置可放宽地址预检
44
HARNESS
5

Configure the composer send shortcut

Add a persisted composer preference that lets users choose whether Enter or Cmd/Ctrl+Enter sends a message, with the other action inserting a newline. Preserve IME composition, default behavior, and Queue/Steer semantics during execution.

Explore challenge ↗

The problem

Add a persisted composer preference that lets users choose whether Enter or Cmd/Ctrl+Enter sends a message, with the other action inserting a newline. Preserve IME composition, default behavior, and Queue/Steer semantics during execution.

Possibly addressed by:Composer Modenter · Composer Keys Setting · Enter Send

Start with these discussions

  • 功能建议：Composer 输入框支持可配置的发送键（Ctrl/Cmd+Enter 发送，Enter 换行）github · ▲4 · 3 comments
  • 把「发送/换行」做成可配置键位：Enter 换行、Ctrl+Enter 发送github · ▲2 · 0 comments
  • [Feature Proposal] Configurable idle Enter behavior: Enter = Send or Newline (Cmd/Ctrl+Enter to send)github · ▲1 · 3 comments
45
HARNESS
4hackernews2

Let users choose models and reasoning per task

Provide task-aware model selection with a provider-organized picker and an independent reasoning-effort control. Make the selection available in Composer and workflow dialogs, with persistence and keyboard-accessible controls.

Explore challenge ↗

The problem

Provide task-aware model selection with a provider-organized picker and an independent reasoning-effort control. Make the selection available in Composer and workflow dialogs, with persistence and keyboard-accessible controls.

Possibly addressed by:Model Selector · Effort · Model Select Plus

Start with these discussions

  • 切换模型的时候能不能别默认恢复推理等级？github · ▲2 · 2 comments
  • [Feature Proposal] Direct reasoning-effort slider beside the composer model selectorgithub · ▲2 · 1 comments
  • [Feature] Provider-first two-step model picker in the composer (plus a direct reasoning-effort slider)
46
HARNESS
4hackernews1

Support Workspace Relocation and Rebinding

Provide a supported workflow to rebind workspaces after folder moves or renames while preserving sessions and updating related paths, configuration, history, metadata, and caches. Safely isolate corrupt or mismatched sessions so relocation does not prevent startup.

Explore challenge ↗

The problem

Provide a supported workflow to rebind workspaces after folder moves or renames while preserving sessions and updating related paths, configuration, history, metadata, and caches. Safely isolate corrupt or mismatched sessions so relocation does not prevent startup.

Possibly addressed by:Session Workspace · Workspace Mover · Session Mover

Start with these discussions

  • Renaming a workspace folder makes its stored sessions fail the identity check, with no supported repairgithub · ▲2 · 2 comments
  • 工作区文件夹绝对位置改变后工作区重绑定功能需求github · ▲2 · 1 comments
47
PLUGIN
2hackernews42

Provide Remote-Control Clients for DSH

Provide polished desktop and mobile clients for connecting to remote DSH servers, monitoring and steering agents, reviewing work, taking over when needed, and returning conveniently to terminal workflows.

Explore challenge ↗

The problem

Provide polished desktop and mobile clients for connecting to remote DSH servers, monitoring and steering agents, reviewing work, taking over when needed, and returning conveniently to terminal workflows.

Possibly addressed by:Mobile Remote · Phone Remote · Remote X

Start with these discussions

  • 桌面端和手机端远程连接服务器上的dsh-web ?github · ▲1 · 1 comments
48
PLUGIN
5

Decouple web search providers from the conversation model

Allow web search to use independently configured providers, credentials, endpoints, and optional auxiliary models, including a zero-LLM mode and reuse of the selected model when needed. Clearly expose provider failures and any search-related model charges, with appropriate safeguards and cost tracking.

Explore challenge ↗

The problem

Allow web search to use independently configured providers, credentials, endpoints, and optional auxiliary models, including a zero-LLM mode and reuse of the selected model when needed. Clearly expose provider failures and any search-related model charges, with appropriate safeguards and cost tracking.

Possibly addressed by:Web Search Diy · Web Search · Websearch

Start with these discussions

  • web_search 在会话模型切到第三方 provider 后仍走 DEEPSEEK_API_KEY 独立计费——披露藏在参考页一行里，且从不说破计费后果github · ▲2 · 4 comments
  • [DSH Desktop] web_search 联网搜索工具静默使用官方 DeepSeek API 并计费，且界面无任何提示github · ▲2 · 2 comments
  • 标题： [Feature Request] web_search 工具应支持不经大模型的纯搜索模式（当前每次搜索都调用付费 DeepSeek API）
49
HARNESS
1hackernews31

Provide a Headless CLI

Expose DeepSeek Harness through a terminal-based, headless CLI that supports developer workflows, automation, coordination, and the core capabilities of the desktop experience.

Explore challenge ↗

The problem

Expose DeepSeek Harness through a terminal-based, headless CLI that supports developer workflows, automation, coordination, and the core capabilities of the desktop experience.

Possibly addressed by:Cli · Dcli · Cli Mode

Start with these discussions

  • DSH应该支持CLIgithub · ▲2 · 4 comments
  • 有没有想过 DeepSeek 这种有个性 比较极客的公司,x · ▲0 · 1 comments
50
HARNESS
4

Implement scoped, persistent agent memory

Provide a backend-agnostic, first-party memory plugin or service with session, agent, tenant, topic, and repository scoping; automatic capture and recall; safe lifecycle handling; provenance, deduplication, bounded injection, and tools for managing facts and lessons. Support importing existing memories and preserving history safely across migrations.

Explore challenge ↗

The problem

Provide a backend-agnostic, first-party memory plugin or service with session, agent, tenant, topic, and repository scoping; automatic capture and recall; safe lifecycle handling; provenance, deduplication, bounded injection, and tools for managing facts and lessons. Support importing existing memories and preserving history safely across migrations.

Possibly addressed by:Agent Memory · Plugin Memory · Hybrid Memory

Start with these discussions

  • dsh-plugin: Native OpenViking memory for DeepSeek Harness (automatic session capture + memory tools + profile injection)github · ▲2 · 5 comments
  • 给 #14 补点料：两个现成记忆实现里踩过的坑github · ▲1 · 4 comments
51
HARNESS
1hackernews91

Implement Policy-Based Automatic Model Routing

Automatically select and route each task to the most suitable model based on capabilities, cost, priority, health, quotas, and fallback policies, including delegation to specialized models.

Explore challenge ↗

The problem

Automatically select and route each task to the most suitable model based on capabilities, cost, priority, health, quotas, and fallback policies, including delegation to specialized models.

Possibly addressed by:Model Router · Model Tier · Model Router

Start with these discussions

  • Design: Policy-driven multi-source quota router for DSHgithub · ▲1 · 0 comments
52
HARNESS
3hackernews2

Stabilize and Dynamically Load Extensions

Provide a documented, version-compatible extension API and loading mechanism that supports persistent third-party plugins, including shared-library loading without recompiling the harness.

Explore challenge ↗

The problem

Provide a documented, version-compatible extension API and loading mechanism that supports persistent third-party plugins, including shared-library loading without recompiling the harness.

Possibly addressed by:Loader · Plugin Extension Center · Plugin Optimization

Start with these discussions

  • Feature Request: Official extension directory for user plugins with stable service injectiongithub · ▲1 · 1 comments
  • [Idea] 文件级自学习回路：把完成的会话蒸馏成跨会话耐久教训（附纯插件实证 @garvel/dsh-reflect）github · ▲1 · 0 comments
  • Plugin API gap on master: settingsNamespace / registerContinuableSetup / deepFreeze missing
53
PLUGIN
2

Add a workspace sandbox cleanup command

Automatically remove temporary Windows sandbox ACLs, deny entries, and Low-integrity labels when a workspace is no longer used, and provide an official command to clean up already-provisioned folders.

Explore challenge ↗

The problem

Automatically remove temporary Windows sandbox ACLs, deny entries, and Low-integrity labels when a workspace is no longer used, and provide an official command to clean up already-provisioned folders.

Possibly addressed by:Workspace Acl Allow · Acl Sandbox Patch · Tool Workspace

Start with these discussions

  • [Bug][Windows] workspace-write leaves a permanent Low integrity label on the project, breaking every other tool used on it — several reports, still unchanged in 0.2.0-rc.2github · ▲7 · 8 comments
  • [Bug] 工作区常驻 Low 完整性标签导致资源管理器把所有文件显示为"来自其他计算机"，且"解除锁定"显示找不到该文件必然失败，移动文件时警告"这些文件可能对你的计算机有害 你的Internet安全设置阻止打开一个或多个文件"github · ▲2 · 1 comments
Last active Sep 29, 2026
2
54
PLUGIN
21

Add a VS Code extension with SSH support

Provide an official VS Code extension for running DeepSeek Harness from the editor sidebar, including connections to remote development systems over SSH.

Explore challenge ↗

The problem

Provide an official VS Code extension for running DeepSeek Harness from the editor sidebar, including connections to remote development systems over SSH.

Possibly addressed by:Vscode Shell · Vsceditor · Visual Studio

Start with these discussions

  • 桌面客户端出来了，vscode扩展什么时候出来啊github · ▲3 · 3 comments
  • 关于vs code插件github · ▲1 · 3 comments
  • @nitin_builds No you are not, i am using cursor so that is essentially vscode. Still using the same extensions etc. The
55
HARNESS
8

Add an Extensible Plugin System

Create a plugin-based core and extend the harness and UI so users can install and create plugins across GUI and CLI environments, with built-in desktop capabilities implemented through the ecosystem.

Explore challenge ↗

The problem

Create a plugin-based core and extend the harness and UI so users can install and create plugins across GUI and CLI environments, with built-in desktop capabilities implemented through the ecosystem.

Possibly addressed by:Deepseek Harness Plugins · Deepseek Harness Plugin · Plugin Smart Config

Start with these discussions

  • @nachooolk @emanueledpt Also do you feel like it would be good if there was a plugin/customisation system so that you cax · ▲0 · 2 comments
  • 你们应该做成 DeepSeek Harness 那样x · ▲0 · 0 comments
  • @DeepSeekHarness can you extend the deepseek harness (including the UI) with plugin?x · ▲0 · 0 comments
Last active Oct 4, 2026
8 authors8 posts · 0 reactions2 comments
56
HUB
hackernews6

Add cross-harness benchmark comparisons

Provide a standardized place to run, publish, and compare agent-harness benchmarks using consistent prompts and models so users can evaluate DSH against alternatives.

Explore challenge ↗

The problem

Provide a standardized place to run, publish, and compare agent-harness benchmarks using consistent prompts and models so users can evaluate DSH against alternatives.

Possibly addressed by:Harness Audit · Plugin Compare · Harness Self Evolution

Start with these discussions

  • Comment on: OpenCode 2.0hackernews · ▲0 · 0 comments
  • Comment on: DeepSeek Harness Desktop for macOS and Windowshackernews · ▲0 · 0 comments
  • Comment on: HarnessTax: How Much Does the Harness Matter for Coding Agents?hackernews · ▲0 · 0 comments
Last active Oct 15, 2026
6 authors6 posts · 0 reactions0 comments
57
PLUGIN
3

Gracefully isolate plugin failures

Add soft-failure handling that records plugin rejection or observer errors and disables or isolates only the affected plugin, allowing the host and healthy plugins to continue. Preserve strict fail-fast behavior where required, while keeping the remaining configuration intact.

Explore challenge ↗

The problem

Add soft-failure handling that records plugin rejection or observer errors and disables or isolates only the affected plugin, allowing the host and healthy plugins to continue. Preserve strict fail-fast behavior where required, while keeping the remaining configuration intact.

Possibly addressed by:Bundle Dedup Guard · Plugin Allcrash · Escape Hatch

Start with these discussions

  • Client plugin-boot failures are swallowed: “The client Loader did not provide an error message” (a plugin's rejected apply() → whole app into recovery mode)github · ▲1 · 4 comments
  • Third-party plugin exception in a session observer takes down the whole hostgithub · ▲1 · 2 comments
  • Loader should tolerate plugin failures: skip the failing entry instead of failing the whole profile
58
HARNESS
3

Add Russian localization

Localize DeepSeek Harness, including its Web UI, for Russian-speaking users with translated strings, correct pluralization, language switching, accessibility metadata, and layout support.

Explore challenge ↗

The problem

Localize DeepSeek Harness, including its Web UI, for Russian-speaking users with translated strings, correct pluralization, language switching, accessibility metadata, and layout support.

Possibly addressed by:Russian Lang · Locale Ru · Deepseek Harness Locale Ru

Start with these discussions

  • Proposal: add Russian (ru) locale to the Web UI (zh/en/ru)github · ▲4 · 4 comments
  • Proposal: Russian (ru) as a third GUI localegithub · ▲1 · 2 comments
  • Добавтее в дипсик харнес русский язык для русскоязычных пользователейgithub · ▲1 · 0 comments
Last active Sep 30, 2026
59
GENERAL
3

Expose model capability metadata

Expose each model’s reasoning-effort support, defaults, and multimodal input metadata so the model picker can display accurate capability indicators and clients can validate selections.

Explore challenge ↗

The problem

Expose each model’s reasoning-effort support, defaults, and multimodal input metadata so the model picker can display accurate capability indicators and clients can validate selections.

Possibly addressed by:Better Reasoning Effort · More Models Thinking Level · Reasoning Effort

Start with these discussions

  • 解决一下这个,自定义提供商默认不能设置思考等级github · ▲3 · 5 comments
  • Expose reasoning effort metadata in `GET /models`github · ▲1 · 1 comments
  • composer 模型下拉：每条候选缺能力标识（思考 effort + 多模态输入）github · ▲1 · 0 comments
Last active Sep 27, 2026
60
HARNESS
1hackernews51

Require confirmation for risky actions

Require explicit approval for destructive, costly, irreversible, or production-impacting actions, including when full-access mode normally disables approval. Support configurable boundaries such as protected paths, spending limits, and destructive-only confirmation.

Explore challenge ↗

The problem

Require explicit approval for destructive, costly, irreversible, or production-impacting actions, including when full-access mode normally disables approval. Support configurable boundaries such as protected paths, spending limits, and destructive-only confirmation.

Possibly addressed by:Agent Guard Dsh · Agent Approval · Always Require Tools Approval

Start with these discussions

← PreviousPage 3 of 13Next →
github · ▲4 · 0 comments
Last active Sep 15, 2026
6 authors6 posts · 27 reactions16 comments
Last active Sep 25, 2026
5 authors5 posts · 9 reactions6 comments
github · ▲1 · 1 comments
Last active Sep 26, 2026
6 authors6 posts · 6 reactions4 comments
  • [Bug] 项目目录迁移后 dsh 启动崩溃（corrupt session log），希望支持会话重新归属github · ▲2 · 1 comments
    • Last active Sep 27, 2026
    5 authors5 posts · 7 reactions4 comments
    win,安卓客户端连接 dsh web
    github · ▲1 · 0 comments
  • Okay so I'm spoiled by ChatGPT Codex Remote Control. What compares in the OSS world?x · ▲0 · 1 comments
    • Last active Oct 2, 2026
    6 authors8 posts · 2 reactions2 comments
    github · ▲2 · 1 comments
    Last active Sep 22, 2026
    5 authors5 posts · 9 reactions8 comments
  • Comment on: DeepSeek Harness Desktop for macOS and Windowshackernews · ▲0 · 0 comments
    • Last active Oct 2, 2026
    5 authors5 posts · 2 reactions5 comments
  • 是否可以支持多租户记忆隔离，提供详细的api文档github · ▲1 · 3 comments
    • Last active Sep 25, 2026
    4 authors4 posts · 5 reactions13 comments
    Comment on: GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price
    hackernews · ▲0 · 0 comments
  • Comment on: >10x More Efficient Pretraininghackernews · ▲0 · 0 comments
    • Last active Sep 29, 2026
    11 authors11 posts · 1 reactions0 comments
    github · ▲1 · 0 comments
    Last active Oct 2, 2026
    5 authors5 posts · 3 reactions1 comments
    authors
    2 posts · 9 reactions
    9 comments
    x · ▲1 · 0 comments
    Last active Sep 30, 2026
    3 authors3 posts · 5 reactions6 comments
    github · ▲1 · 1 comments
    Last active Oct 1, 2026
    3 authors3 posts · 3 reactions7 comments
    2 authors3 posts · 6 reactions6 comments
    3 authors3 posts · 5 reactions6 comments
    [Feature] danger-full-access + approval=never 组合下，不可逆删除缺少最终确认（附真实事故）
    github · ▲1 · 0 comments
  • Comment on: Claude Opus 5.5hackernews · ▲0 · 0 comments
  • Comment on: GPT-6 Sol and Lunahackernews · ▲0 · 0 comments
    • Last active Oct 2, 2026
    7 authors7 posts · 1 reactions0 comments