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.

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 261–280 of 289 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
261
HARNESS
hackernews5

Enforce Output Style with Stop Hooks

Allow configurable stop hooks to detect unwanted code comments, excessive verbosity, or prohibited words and trigger regeneration. The checks should suppress conversational or redundant comments while preserving useful implementation comments.

Explore challenge ↗

The problem

Allow configurable stop hooks to detect unwanted code comments, excessive verbosity, or prohibited words and trigger regeneration. The checks should suppress conversational or redundant comments while preserving useful implementation comments.

Possibly addressed by:Inline Comments · Improved Inline Edit · Agent Guard Dsh

Start with these discussions

  • Comment on: Claude Fable 5.1 and Claude Mythos 5.1hackernews · ▲0 · 0 comments
  • Comment on: The Harness Is the Thinghackernews · ▲0 · 0 comments
  • Comment on: Why does Opus 5 feel worse to work with?hackernews · ▲0 · 0 comments
Last active Sep 1, 2026
5 authors5 posts · 0 reactions0 comments
262
HUB
hackernews2

Rank and showcase popular setups

Aggregate community-submitted setup data and rank configurations by popularity and reviews. Highlight top-quality setups to help newcomers discover and adopt useful tools.

Explore challenge ↗

The problem

Aggregate community-submitted setup data and rank configurations by popularity and reviews. Highlight top-quality setups to help newcomers discover and adopt useful tools.

Possibly addressed by:Weekly Hot · Recommend · Plugin Recommend

Start with these discussions

  • Comment on: Show HN: Share your AI Setup, Learn from othershackernews · ▲0 · 0 comments
  • Comment on: Show HN: Share your AI Setup, Learn from othershackernews · ▲0 · 0 comments
Last active Sep 22, 2026
2 authors2 posts · 0 reactions0 comments
263
HARNESS
11

Manage agent sessions through the Web GUI

Provide a graphical interface for managing agent sessions, including inspecting attached long-term memory, closing or detaching live sessions, releasing write leases, and viewing lease ownership details. This should eliminate reliance on CLI tools for routine session management.

Explore challenge ↗

The problem

Provide a graphical interface for managing agent sessions, including inspecting attached long-term memory, closing or detaching live sessions, releasing write leases, and viewing lease ownership details. This should eliminate reliance on CLI tools for routine session management.

Possibly addressed by:Session Hub · Session Manager · More Session Operations

Start with these discussions

  • [Feature] Web GUI 增加「关闭会话」动作 + 写租约持有者提示(多窗口/多 profile 并发体验)github · ▲1 · 1 comments
264
HARNESS
1hackernews1

Expose real-time agent progress and step counts

Expose an agent’s current progress while it is running, including the number of steps taken, so the main agent can monitor long-running subagents and inspect their progress during or after execution.

Explore challenge ↗

The problem

Expose an agent’s current progress while it is running, including the number of steps taken, so the main agent can monitor long-running subagents and inspect their progress during or after execution.

Possibly addressed by:Subagent Progress · Subagent Steer · Background Agents

Start with these discussions

  • 【功能】对于子 agent 的改进github · ▲1 · 0 comments
  • Comment on: Gemini 3.8 Flash and 3.8 Flash Cyberhackernews · ▲0 · 0 comments
Last active Sep 9, 2026
2 authors
265
HARNESS
3

Enable Codex-to-DeepSeek Harness message delivery

Enable bidirectional communication so Codex can send information directly to DeepSeek Harness, eliminating manual copy-and-paste between the two systems.

Explore challenge ↗

The problem

Enable bidirectional communication so Codex can send information directly to DeepSeek Harness, eliminating manual copy-and-paste between the two systems.

Possibly addressed by:Codex · Codex Community · Embedded Codex

Start with these discussions

  • i might even try to add on the deepseek harness which was released if i can get it to connect to codex i will be runningx · ▲0 · 0 comments
  • @thsottiaux Tibo,我今天意外发现Deepseek harness内的子线程的可以直接将信息发送给codex内的sol主线程,通过codex cli queue。这样我就不需要手动粘贴来传递信息了,不过怎么才能让codex发送x · ▲0 · 0 comments
  • 我今天意外发现Deepseek harness内的子线程的可以直接将信息发送给codex内的sol主线程,通过codex cli queue。这样我就不需要手动粘贴来传递信息了,不过怎么才能让codex发送信息给Deepseek harnex · ▲0 · 0 comments
Last active Sep 18, 2026
2 authors3 posts · 0 reactions0 comments
266
HARNESS
hackernews2

Support gVisor and Kata/Cloud-Hypervisor Sandboxes

Add support for running agents in gVisor or container sandboxes backed by Kata and Cloud Hypervisor, providing stronger isolation options.

Explore challenge ↗

The problem

Add support for running agents in gVisor or container sandboxes backed by Kata and Cloud Hypervisor, providing stronger isolation options.

Possibly addressed by:Sandbox Temp Guard · Plugin Vm Sandbox · Sandbox

Start with these discussions

  • Comment on: Show HN: Pi pod – Run your pi coding agent in sandboxes on your own serverhackernews · ▲0 · 0 comments
  • Comment on: Show HN: Pi pod – Run your pi coding agent in sandboxes on your own serverhackernews · ▲0 · 0 comments
Last active Oct 4, 2026
1 authors2 posts · 0 reactions0 comments
267
HARNESS
hackernews2

Improve resumable session management

Provide session renaming that keeps history selectable for resumption, along with the requested forking and keep/undo workflows. Renaming the underlying history folders should remain consistent with the session identity.

Explore challenge ↗

The problem

Provide session renaming that keeps history selectable for resumption, along with the requested forking and keep/undo workflows. Renaming the underlying history folders should remain consistent with the session identity.

Possibly addressed by:Session Rename · More Session Operations · Session Mover

Start with these discussions

  • Comment on: Claude Code now reads AGENTS.md if there is no Claude.mdhackernews · ▲0 · 0 comments
  • Comment on: Our decision on Cursor following its acquisition by SpaceXhackernews · ▲0 · 0 comments
Last active Sep 19, 2026
2 authors2 posts · 0 reactions0 comments
268
HARNESS
hackernews2

Use platform-standard application directories

Store settings, preferences, and application data in each operating system’s standard per-user locations instead of scattering files throughout the home directory, including appropriate macOS directories.

Explore challenge ↗

The problem

Store settings, preferences, and application data in each operating system’s standard per-user locations instead of scattering files throughout the home directory, including appropriate macOS directories.

Start with these discussions

  • Comment on: DeepSeek Harness Desktop for macOS and Windowshackernews · ▲0 · 0 comments
  • Comment on: DeepSeek Harness Desktop for macOS and Windowshackernews · ▲0 · 0 comments
Last active Oct 3, 2026
1 authors2 posts · 0 reactions0 comments
269
HARNESS
hackernews11

Cache and reuse prompt-processing slots

Add native prompt/prefix caching that saves prompt-processing slots, optionally in a RAM disk, and reuses them across forked conversations, chat windows, subagents, and parallel requests to avoid duplicate prefill work and reduce latency.

Explore challenge ↗

The problem

Add native prompt/prefix caching that saves prompt-processing slots, optionally in a RAM disk, and reuses them across forked conversations, chat windows, subagents, and parallel requests to avoid duplicate prefill work and reduce latency.

Possibly addressed by:Granular Prompt · Prompt Injector · Prompt Stash

Start with these discussions

  • Prompt processing takes 21 seconds, so reading it from a RAMDISK should reduce it to 1-2 secondsx · ▲0 · 0 comments
  • Comment on: Speculative Decoding in vLLM on AMD GPUshackernews · ▲0 · 0 comments
Last active Sep 18, 2026
2 authors2 posts · 0 reactions0 comments
270
HARNESS
2

Expose transparent agent activity and diagnostics in UIs

Provide UI views alongside or within TUIs that expose live agent activity, workbench logs, and diagnostic information such as blank responses, OOM errors, and context truncation.

Explore challenge ↗

The problem

Provide UI views alongside or within TUIs that expose live agent activity, workbench logs, and diagnostic information such as blank responses, OOM errors, and context truncation.

Possibly addressed by:Workbench Ui · Deepseek Harness Tui · Deepseek Harness Chat Ui

Start with these discussions

  • @dhh True, this was actually one of the reasons when I saw the "DeepSeek harness" I was like, I wish we had that in allx · ▲1 · 0 comments
  • DeepSeek Harness returns blank after a long task? Don't swap the API key first. Open the workbench log: look for OOM orx · ▲0 · 0 comments
Last active Sep 5, 2026
2 authors2 posts · 1 reactions0 comments
271
PLUGIN
2

Persist corrections as reviewed project knowledge

Provide a workflow that captures corrections in an inbox, supports review, and turns approved items into durable project knowledge so users do not need to repeat them. The workflow should integrate with DeepSeek Harness through MCP.

Explore challenge ↗

The problem

Provide a workflow that captures corrections in an inbox, supports review, and turns approved items into durable project knowledge so users do not need to repeat them. The workflow should integrate with DeepSeek Harness through MCP.

Possibly addressed by:Obsidian Inbox · Codex Workflow · Evidence First Knowledge Work

Start with these discussions

  • The workflow I want is simple:x · ▲0 · 0 comments
  • @tianyi @DeepSeekHarness The workflow I want is simple:x · ▲0 · 0 comments
Last active Oct 2, 2026
1 authors2 posts · 0 reactions0 comments
272
HARNESS
1hackernews1

Add interactive diff-based edit review

Show each proposed file change as an editable diff, require approval before proceeding, and let users provide reasons when rejecting changes.

Explore challenge ↗

The problem

Show each proposed file change as an editable diff, require approval before proceeding, and let users provide reasons when rejecting changes.

Possibly addressed by:Approval Diff · Edit Approval · Diff Approval

Start with these discussions

  • 回绝理由与差异预览支持github · ▲1 · 0 comments
  • Comment on: GLM-5.3: Frontier coding with emergent cyber capabilitieshackernews · ▲0 · 0 comments
Last active Sep 4, 2026
2 authors2 posts · 1 reactions0 comments
273
HARNESS
hackernews2

Support metric units

Provide metric-unit support throughout the harness, including meters for measurement and monitoring capabilities.

Explore challenge ↗

The problem

Provide metric-unit support throughout the harness, including meters for measurement and monitoring capabilities.

Possibly addressed by:Usage Meter Harness · Units · Meter

Start with these discussions

  • Comment on: Ask HN: What are you working on? (September 2026)hackernews · ▲0 · 0 comments
  • Comment on: Ask HN: What are you working on? (August 2026)hackernews · ▲0 · 0 comments
Last active Sep 14, 2026
2 authors2 posts · 0 reactions0 comments
274
HARNESS
2

Recognize the published Typert protocol in external plugins

Allow the harness to recognize declarations from the exact @deepseek-ai/dsh-typert-protocol package in independently published external plugins, so their typed Remote methods are discovered and Remote artifacts are generated without accepting unrelated lookalike packages.

Explore challenge ↗

The problem

Allow the harness to recognize declarations from the exact @deepseek-ai/dsh-typert-protocol package in independently published external plugins, so their typed Remote methods are discovered and Remote artifacts are generated without accepting unrelated lookalike packages.

Possibly addressed by:Writing Remote · Plugin Template · Plugin Design

Start with these discussions

  • Proposal: recognize official Typert protocol metadata in third-party plugin packagesgithub · ▲1 · 0 comments
  • Proposal: recognize the published Typert protocol in external pluginsgithub · ▲1 · 0 comments
Last active Sep 5, 2026
1
275
HARNESS
hackernews2

Add a safety and morality framework for agents

Provide shared ethical and safety constraints that guide agent behavior and prevent unauthorized actions such as escaping containers, accessing host credentials, or making unsafe external changes.

Explore challenge ↗

The problem

Provide shared ethical and safety constraints that guide agent behavior and prevent unauthorized actions such as escaping containers, accessing host credentials, or making unsafe external changes.

Start with these discussions

  • Comment on: Why are AI agents lying, cheating and coordinating?hackernews · ▲0 · 0 comments
  • Comment on: We must pace the frontierhackernews · ▲0 · 0 comments
Last active Sep 14, 2026
2 authors2 posts · 0 reactions0 comments
276
HARNESS
3

Integrate Codex and Claude Code with DeepSeek Harness

Enable DeepSeek Harness to connect to and run Codex and Claude Code as integrated coding agents, including Claude Code access where the current Command Goat subscription setup does not support it.

Explore challenge ↗

The problem

Enable DeepSeek Harness to connect to and run Codex and Claude Code as integrated coding agents, including Claude Code access where the current Command Goat subscription setup does not support it.

Possibly addressed by:Claude Bridge · Relay Dsh Plugin Claude · Claude To Dsh

Start with these discussions

  • 兄弟们,刚才试了, command goat 订阅 在Deepseek harness 里面跑Deepseek v4.1 flash ,x · ▲0 · 0 comments
  • @aaronqi 哈哈,我是小扒菜,不客气,刚才试了, command goat 订阅 在Deepseek harness 里面跑Deepseek v4.1 flash ,x · ▲0 · 0 comments
  • @Airdrop_Guard 建议 deepseek harness➕codex 或者是 Claude codex · ▲0 · 0 comments
Last active Sep 10, 2026
2 authors3 posts · 0 reactions0 comments
277
HARNESS
1hackernews1

Add a persistent ACP server mode

Provide an ACP subcommand or server/RPC mode with a persistent endpoint for external tools and supervisors to discover, control, and maintain stateful DSH or coding-agent sessions.

Explore challenge ↗

The problem

Provide an ACP subcommand or server/RPC mode with a persistent endpoint for external tools and supervisors to discover, control, and maintain stateful DSH or coding-agent sessions.

Possibly addressed by:Acp · Acp · Acp Server

Start with these discussions

  • feat: Missing ACP/Multica daemon long-running protocol support (blocking integration)github · ▲1 · 1 comments
  • Comment on: fx :Tiny, open, native coding agent.hackernews · ▲0 · 0 comments
Last active Aug 24, 2026
2 authors2 posts · 1 reactions1 comments
278
HUB
2

Provide a polished Oh My DSH community plugin bundle

Create a polished, community-maintained Oh My DSH-style bundle that makes DSH easy to set up and use through compatible, well-integrated plugins.

Explore challenge ↗

The problem

Create a polished, community-maintained Oh My DSH-style bundle that makes DSH easy to set up and use through compatible, well-integrated plugins.

Possibly addressed by:Custom Plugin · Plugin Kit · Plugin Security Review

Start with these discussions

  • Pi Agent 还是 DeepSeek Harness?x · ▲1 · 0 comments
  • Pi Agent 还是 DeepSeek Harness?x · ▲0 · 0 comments
Last active Sep 4, 2026
1 authors2 posts · 1 reactions0 comments
279
HARNESS
hackernews4

Run model inference locally

Allow agents to load and run locally hosted or specialized models directly on the device, without depending on Vercel-hosted inference or requiring a Vercel account.

Explore challenge ↗

The problem

Allow agents to load and run locally hosted or specialized models directly on the device, without depending on Vercel-hosted inference or requiring a Vercel account.

Possibly addressed by:Local Ai · Local Models · Llm Qwen Local

Start with these discussions

  • Comment on: fx :Tiny, open, native coding agent.hackernews · ▲0 · 0 comments
  • Comment on: DeepSeek V4 Pro 0813hackernews · ▲0 · 0 comments
  • Comment on: Qwen3.8 Max now ranked as the best overall model by agentic indexhackernews · ▲0 · 0 comments
Last active Aug 19, 2026
4 authors4 posts · 0 reactions0 comments
280
HARNESS
hackernews3

Record and expose complete agent traces

Persist append-only, inspectable records of every agent run, including assembled prompts, model inputs and outputs, tool activity, scheduling, context injections, and thought traces. Support searching, inspection, resumption, forking, and replay so users can examine the agent’s reasoning process.

Explore challenge ↗

The problem

Persist append-only, inspectable records of every agent run, including assembled prompts, model inputs and outputs, tool activity, scheduling, context injections, and thought traces. Support searching, inspection, resumption, forking, and replay so users can examine the agent’s reasoning process.

Possibly addressed by:Agent Run Logger · Trace Collapse · Replay

Start with these discussions

  • Comment on: Claude Fable 5.1 and Claude Mythos 5.1hackernews · ▲0 · 0 comments
  • Comment on: DeepSeek Harness developer previewhackernews · ▲0 · 0 comments
  • Comment on: Claude Code: Starting August 14, auto mode will be the default permission modehackernews · ▲0 · 0 comments
Last active Sep 1, 2026
2 authors3 posts · 0 reactions0 comments
← PreviousPage 14 of 15Next →
Hermes Agent replies feel amnesiac after a restart? Open the session panel and check whether long-term memory is attache
x · ▲0 · 0 comments
Last active Sep 5, 2026
2 authors2 posts · 1 reactions1 comments
2 posts · 1 reactions
0 comments
authors
2 posts · 2 reactions
0 comments