Plugins
Durchsuche, filtere und installiere DeepSeek-Harness-Plugins.
466 Plugins gefunden
dsh-max-reasoning-effort
devzer0o0/dsh-max-reasoning-effort
Pins reasoning effort to the highest level every model offers, and keeps it pinned for agents created after the plugin loads.
dsh-think-flow
cv588888888888888888888ju/dsh-think-flow
Chinese add-on fork of the dsh-think-flow sidebar chain-of-thought panel: while upstream folds each session's live reasoning into turn → step groups, this fork sends each step to a separate, configurable model (titleProvider/titleModel) and shows an 8–18 character Chinese title plus a 20–60 character Chinese explanation under every step, and a one-line Chinese title per turn. The main model is untouched — no system-prompt injection, no extra tools, no context changes; the plug-in just makes its own request. autoTitlesOn arms title generation as soon as a panel connects, and batches over 10 steps are split so long turns (100+ steps) no longer come back empty; steps that only call a tool (no reasoning) are summarised from the call itself, and a failed segment is retried up to three times with backoff.
dsh-plugin-sandbox-mode
wth373737/dsh-plugin-sandbox-mode
A separate agent mode for DeepSeek Harness: reasoning goes into a sandbox document, every action is written there first and then executed by reading it back, and the final answer is printed from that document. Direct tool calls are blocked per preset, and the output also lands as a file card. Deliberately more expensive, and the mode picker says so.
dsh-session-rescue
macilydots/dsh-session-rescue
Session reliability kit: detects reasoning repetition loops and cancels the step, retries reasoning-only empty turns, records raw upstream streams on demand, hands off to a clean session after repeated degeneration, and reloads the renderer with Ctrl+Shift+R.
dsh-qwen-reasoning
jasocheung/dsh-qwen-reasoning
Auto-enhances manually-declared Qwen models with the full reasoning-effort levels (off/minimal/low/medium/high/xhigh/max) and defaults to max. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-prompt-smith
cyberpzy/dsh-prompt-smith
Adds a sparkle button to the DSH composer's tool row that rewrites the current draft into a structured prompt through one auxiliary call over the profile's own LLM route, writes the result back into the input (Ctrl+Z restores the original), fences the draft against prompt injection, and never touches the session log; a settings page selects provider, model and reasoning effort from the host's own model catalog with a test call that reports unsupported effort levels, edits the built-in system prompt (persisted through the host config editor), and tunes temperature, max tokens and max draft length. Zero dependencies, no build step.
dsh-dragon-maid-pet
xnwang1999/dsh-dragon-maid-pet
A desk pet for the DSH web GUI whose illustration follows the current model's reasoning effort across five stages, taken from one of three art sets that auto-select for Claude and DeepSeek models. Draggable and collapsible, with a custom image per stage, and it also fills in reasoning-effort levels for custom model providers in Settings → Models.
dsh-thinking-level-zh
yulongjiang1997/dsh-thinking-level-zh
Shows model reasoning effort levels in Chinese everywhere in the conversation dialog, with an editable level-to-label mapping in Settings.
dsh-share-session
fruneng/dsh-share-session
Generates a shareable long PNG from selected turns of a DSH session, with optional reasoning and tool activity.
dsh-handoff-at-75
1502074527-svg/dsh-handoff-at-75
Hands off a conversation when context usage reaches a configured threshold: writes a handoff document into the project and continues the task in a new session in the same workspace, with the new title carrying a -N suffix, an optional translucent four-color status tint on sidebar session rows, and a configurable model + reasoning effort for the handoff conversation (also selectable in a prompt at handoff time).
dsh-connect-officeAce
hxgh776/dsh-connect-officeace
Connects a locally signed-in Huawei OfficeAce desktop account to DSH as an OpenAI-compatible chat provider that reuses the stored login instead of an API key, with streaming, reasoning and tool-call passthrough.
dsh-nvidia-cloud-provider
ai-workspace-lab/dsh-nvidia-cloud-provider
NVIDIA NIM cloud provider for DeepSeek Harness with model discovery, tool calling and reasoning stream.
dsh-channel-feishu
poy-coder/dsh-channel-feishu
Feishu/Lark channel for DeepSeek Harness: one chat bound to one DSH session, streaming CardKit answers, the native reasoning process, interactive question cards, attachments, and reconnect/dedupe hardening for unattended use. Inbound is off until you turn it on; once on, an empty allowedUsers lets the first sender become the owner, and channel sessions run under the workspace-write preset unless you change it.
dsh-subagent-route-args
gv-megumi/dsh-subagent-route-args
Treats empty-string or whitespace-only provider, model and reasoning_effort arguments on the subagent, subagent_fork and list_subagent_models tools as omitted, so DSH falls back to the parent route instead of rejecting the call.
dsh-connect-comate
tsd619124/dsh-connect-comate
Registers the Comate (Zulu) models of a locally signed-in Comate CLI as a DSH provider, supervising a loopback `zulu serve` and translating its SSE into the DSH stream. One Comate conversation is reused per DSH session, an interruption that waits for a choice becomes a visible question you can answer in the next turn, the reasoning effort and image input the roster advertises are actually forwarded, a Settings section trims the model picker to the models you tick, and a `comate_usage` tool reports real token usage plus the daily credit. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-quick-model-switch
caidows/dsh-quick-model-switch
Codex-style composer model picker: a card showing the current model and a draggable slider that sets reasoning effort, with one stop per level the model advertises and an Off stop when the provider supports it. The slider snaps only to whole stops and updates the effort label while dragging. Up to three configurable model+effort pairs sit beside it as one-click quick-switch chips, configured in the plugin's own popover. Model and effort switches still go through the session model directory, so they stay in sync with the rest of the composer.
dsh-auto-team
maliang/dsh-auto-team
Turns the current session into a team lead: splits a goal into a dependency graph of tasks, assembles a team from 11 built-in roles (each optionally bound to its own provider, model and reasoning effort), and dispatches ready tasks concurrently into their own sub-sessions. A task is done only after its acceptance command runs through a receipt that records the exit code and output tail; a live panel reports duration and token totals per run, task and member.
dsh-pipem
baimacao/dsh-pipem
Error ledger and long-term memory for DSH: records each mistake as a searchable guard, injects it before you retry, auto-reminds on tool failure, detects reasoning loops, and ships bash plus a persona/journal system.
dsh-dafeiyu-zh-thinking
sblzc/dsh-dafeiyu-zh-thinking
Appends a one-line Chinese-reasoning reminder to the tail of every prepared request, because a long session drifts back to English when its own earlier reasoning outvotes the system prompt.
dsh-connect-loomy
hxgh776/dsh-connect-loomy
Connects the locally signed-in Loomy (iFlytek) desktop account to DeepSeek Harness: chat models with reasoning and tool calls executed locally by DSH, plus a read-only credits dashboard (total balance, daily allowance, available balance). Reuses the Loomy app's own session file without copying or storing credentials; doctor and probe CLIs included. Windows x64.
online_model_discovery
dhllib/online_model_discovery
Adds live model discovery and per-model reasoning-effort editing for third-party pi-ai providers.
dsh-team-switch-place
zrrrrrrr77/dsh-team-switch-place
Puts the Agent Teams switch above the model and reasoning rows in the composer menu, out of mis-tap range. UI-only, no tokens; Tab lands on the reasoning row.
dsh-model-fallback
carlcn2003/dsh-model-fallback
Primary and multi-level backup model failover router for DSH, with cross-provider reasoning-effort clamping, exponential cooldown, per-turn recovery, and a settings panel.
dsh-composer-model-controls
timxiang-948/dsh-composer-model-controls
Reasoning-effort control for the DSH composer — an adaptive effort slider whose track and composer border tint with the level, plus Shift+Tab cycling; the built-in model menu stays untouched.