Skip to main content

Plugins

Browse, filter, and install DeepSeek-Harness plugins.

15886 plugins found

S

dsh-plugin-safe-route

sshooting/dsh-plugin-safe-route

Read-only DSH command that audits which model routes a session used and flags calls outside a configurable allowlist.

02 days agoSecurity & PermissionsMIT
M

dsh-credential-vault

mysterycode-liwen/dsh-credential-vault

A sidebar credential vault for DeepSeek Harness that stores accounts, passwords and API keys locally, with an access lock and an agent read tool. Secrets are masked by default.

021 hours agoSecurity & PermissionsMIT
Y

dsh-dev-tools

yu-bobo/dsh-dev-tools

Sidebar developer utilities for JSON, timestamp, UUID, hash, AES, Base64, Diff, and model-backed translate, with pinable tabs and agent tools.

03 days agoDev & Plugin ToolsMIT
S

dsh-web-updater

shawthua/dsh-web-updater

Update a web-only DSH host from its own web GUI on Windows: read the nightly feed, download the installer, verify its sha512, install it with a detached script, and start the host again.

03 days agoDev & Plugin ToolsMIT
O

dsh-plugin-lab

oseas031/dsh-plugin-lab

Zero-restart acceptance lab for DSH plugins: boots a real cordis Loader and ToolsService in a subprocess, drives the mounted plugin through scenario scripts, and reports PASS/FAIL per check with an overall verdict.

02 days agoDev & Plugin ToolsMIT
K

dsh-artel

ksolodev/dsh-artel

Runs a development task in an isolated copy of the project: Solo runs one agent, Swarm runs three agents in the same copy that message each other via send_message/list_agents, Conductor adds a lead that hands out the shares, answers stuck agents and accepts the work as a whole (up to two rework rounds), and Fusion asks three models the same question in parallel in a read-only sandbox, has a judge model merge their answers into one decision and lets one implementer do the work; the resulting diff is shown per file and only the files you approve are applied. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.

0yesterdayDev & Plugin ToolsMIT
Z

dsh-imessage

zhou200310/dsh-imessage

DSH 0.2.2 adaptation of photon-hq/dsh-imessage: the same Photon-hosted two-way iMessage channel, fixed for the 0.2.2 host line — settings registration, session-persistence reads, message source kind, listener auto-restart after a cold boot, and workspace attachment for plugin-created sessions — with attachments carried both ways. The setup page is fully localized in Chinese.

02 days agoIntegrations & RemoteApache-2.0
X

dsh-message-hub

xiafan68/dsh-message-hub

Brings RSS feeds, web pages and GitHub releases into a three-pane panel and exposes them to the agent as hub_* tools; the bundled Python collector starts on demand and refreshes sources on a timer.

07 days agoIntegrations & RemoteMIT
N

agent-bridge (dsh)

nia-teams/agent-bridge

Connects DeepSeek Harness agents to Nia Teams group chats with two-way messages, per-chat sessions, and human approval cards.

02 days agoIntegrations & RemoteMIT
M

dsh-session-notify

mloluyu/dsh-session-notify

Quick reminder: DSH sends you a system notification when a task finishes, when your approval is needed, or when a question is waiting for your answer.

02 days agoIntegrations & RemoteMIT
H

dsh-system-notifications

hexiao160705-dev/dsh-system-notifications

Native OS notifications for DeepSeek Harness when a turn finishes, an approval is requested, or a question or plan review is waiting, with independent switches, a test button, and click-to-focus.

07 days agoIntegrations & RemoteMIT
Z

dsh-zentao-buglens

zhan1024/dsh-zentao-buglens

BugLens turns annotated screenshots and page evidence into guided bug reports, suggests frontend/backend triage and recent assignees, and creates ZenTao bugs with attached materials.

05 days agoWorkflow & AutomationMIT
X

dsh-task-queue

xuexue61/dsh-task-queue

Task queue as a conversation view: per-project lists with priorities and a per-chat scope, one-click send of an item into the current conversation, per-conversation auto-run by priority, and pending/running/done/failed progress for every item.

03 days agoWorkflow & AutomationMIT
V

dsh-TradingAgents (npm)

vicfei/dsh-tradingagents

Multi-agent investment committee for DSH - analyst, bull, bear, risk and PM debate where every claim carries an evidence tag, abstention (PASS) is a first-class verdict, Buffett/Munger/Lynch/Wood guest seats critique the debate itself, and zero-key market data (CBOE quote/K-line/options-IV + SEC 10-K + nasdaq news, Yahoo-free) feeds an auditable IC-MEMO - 6 to 10 model calls per session.

03 days agoWorkflow & AutomationMIT
Q

dsh-bili-banghui

qisisiqi/dsh-bili-banghui

Tired of the daft replies on Bilibili? This plugin answers them for you: it groups the replies you got by person and thread into conversations, lines up the original comment, their reply and yours on one screen, and when you tap the AI reply button it first reads whether the other person is agreeing, chatting or picking a fight (it will not argue back at someone friendly), then offers three options you can edit into your own voice. Nothing goes out until you confirm, and every send is logged. Sign-in, persona and the model API key live under the avatar menu in the top-right corner; credentials stay in your local data folder. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.

03 days agoSessions & MessagesMIT
K

codex-dsh-dispatch

kazeliu/codex-dsh-dispatch

Lets Codex dispatch tasks to DeepSeek Harness, wait for results, and follow up in the same DSH session.

03 days agoWorkflow & Automation
E

dsh-product-workbench

ethanyuan802/dsh-product-workbench

Provides product management workflows, knowledge search, and editable Office reports in DeepSeek Harness.

05 days agoWorkflow & AutomationMIT
T

figureatlas

tardisbooo/figureatlas

Figure catalog of 300 verified types for papers, slides, posters, and video frames. Agents pick a figure id and export HTML or SVG.

04 days agoSkills
P

dsh-workbuddy-skills

panweizong/dsh-workbuddy-skills

Skills importer for DSH: scans the WorkBuddy skills folder, normalizes and de-duplicates skill ids, then copies the skills you pick into the DSH skills directory from a Settings panel. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.

03 days agoSkillsMIT
J

dsh-poe-cn-guide

jayf4n/dsh-poe-cn-guide

Translation skill for Path of Exile build guides that turns English league-starter write-ups into Chinese with poedb.tw terminology, shipping a 3,134-term glossary and real gem socket colours.

05 days agoSkills
F

endo-decision-workbench

flying-ovaltine/endo-decision-workbench

Endocrinology decision workbench for frontline endocrinologists. A differential table ranked on two axes (likelihood x harm if missed) where every row must carry counter-evidence, one of six evidence grades on each recommendation, branching workup paths and a treatment plan; emergency red flags collapse everything into three lines. The clinical rules ship as the skill endocrinology-decision-team.

04 days agoSkillsMIT
F

chronic-care-workbench

flying-ovaltine/chronic-care-workbench

Patient-facing chronic care workbench. Blood pressure, glucose, weight, waist, uric acid and eGFR are recorded with an A/B/C reliability grade; a medication timeline separates doctor-ordered from self-adjusted doses; self-contained SVG trend charts draw medication events as vertical lines (self-adjusted in red dashes). Red-flag symptoms pin a red banner and collapse the rest, women pass a pregnancy gate first, and the physician-guidance sentence is always pinned. Rules ship as the skill chronic-care-team.

04 days agoSkillsMIT
E

dsh-andrej-karpathy

echo-ix/dsh-andrej-karpathy

Adds an on-demand coding-guidelines skill covering assumptions and tradeoffs, simplicity, surgical changes, and verifiable goals, adapted from Andrej Karpathy's notes on LLM coding pitfalls.

05 days agoSkills
C

dsh-skill-eval

carlslin/dsh-skill-eval

Scores every skill 0-100 from real execution telemetry (load success, follow-up tool errors, user corrections, ignored loads and cost) with 95% confidence intervals, turns that evidence into fixes the agent can act on, and copies one skill's current issues from the board as a ready-to-paste Markdown repair task.

02 days agoSkillsMIT