Plugin
Sfoglia, filtra e installa i plugin di DeepSeek-Harness.
315 plugin trovati
madoka (dsh-magica-rune)
meiyukichan/madoka
Your harness, your rules: tailor dsh to whatever the current job needs by switching any tool or skill on or off — globally, or for one workspace only — from a single Settings page. A disabled ability leaves both the model tool list and the system prompt, so the model stops seeing what it must not use: fewer tokens spent, no ghost calls. Every invocation is still refused at runtime as a second line of defence, and the change lands in the running conversation without a restart. Search by tool name or category, flip a whole category with one checkbox (built-in tools still ask for confirmation), and trust the abilities dsh itself depends on to stay protected. Workspace rules layer on top of the global ones, so the same harness can be locked down for one repo and wide open in another.
dsh-safety-memory
c505016310-maker/dsh-safety-memory
Injects live-editable safety rules and a project-memory workflow into every DSH conversation, and auto-reads the project's MEMO.md.
dsh-session-manager
blindcat-10835/dsh-session-manager
A session manager panel for the DSH Web GUI: sessions are grouped by status (seven states), time bucket or tag, with a group order you can drag into place and a tag vocabulary you can create, rename and delete. Tag rules suggest tags from keywords, AI batch rename and AI tagging always preview before writing and run as cancellable background jobs, and the panel reuses the native sidebar context menu (appending a tag submenu) plus the native row seats to show a status dot and tag count on every sidebar row.
dsh-inbox-planner
char-aaa/dsh-inbox-planner
Turns new mail from every dsh-email account into dated to-dos: drops codes, alerts and newsletters without a model call, reads attached .ics invitations directly, extracts actions and deadlines with the DSH default model (rules fallback, several mails in parallel), skips CC-only mail, dedupes across mailboxes, flags time clashes, and moves or closes a to-do when a later mail reschedules or cancels it. Items go to a built-in to-do panel with a month calendar (adapted from dsh-sidebar-todo, same data file), optionally mirrored to dsh-calendar. Reminders before timed items and a daily brief light a red dot on the To-do icon; the panel header has refresh, an activity log with ignore-sender, and .ics export.
ZJU-mooc-assistant
whoisyanhaoooo/zju-mooc-assistant
A DeepSeek Harness plugin for ZJU students on icourse163 that plays course videos, handles quizzes and exams, posts course discussions, and shows live task progress. It drives a local Edge or Chrome window signed in to your own icourse163 account (slider verification is yours to complete); with Remember password ticked the password is saved in $DSH_HOME/ZJU-mooc-assistant/account.json (mode 0600), and automated quizzes and posts are subject to your course's and the platform's rules.
dsh-session-profiler
jkzing/dsh-session-profiler
Adds an analysis tab to each session that draws it as a timeline: swim-lane tracks by category, a flame graph, and per-category and per-tool timing statistics for model calls, tool calls and background processes. Running sessions refresh live, and extra lanes can be defined with regex rules in the plugin config.
dsh-nanshan-market-publish
southm/dsh-nanshan-market-publish
Scans DSH components under development on this machine (or folders you add) and walks you through two independent tracks from the official contributing.md: pushing the code to GitHub, and opening the market listing PR. It checks the dsh.bundle manifest, the one-day repository age, the dsh-plugin topic, the category and the description rules, then writes data/plugins/<owner>__<repo>.yml. Open Nansh. Its HTTP routes have no DSH access check, and they can push code and open pull requests on GitHub with the token you give it.
dsh-superagent-wechat-source
luckyxiang6688/dsh-superagent-wechat-source
Personal WeChat DMs as a Super-agent Source over the iLink ClawBot channel: bind a WeChat bot by scanning a QR code in the plugin's settings page, then every DM is adjudicated by the intake rules — accepted ones become tickets and post an acknowledgement plus the final result back into the same chat, the rest are recorded as events without a ticket. Text only, one-to-one chats. Only WeChat users on the superagent policy's allow-list are assessed; the list is empty by default, so every DM is ignored until you add your own ilink_user_id.
dsh-plugins (reminders)
genesi599/dsh-plugins
Runs allowlisted Python hook scripts on session events and injects the text they return into the next turn, so standing rules live in editable .py files instead of the prompt. Needs Python on PATH; it only runs scripts whose names are on its built-in allowlist.
dsh-ponytail
yanweiyi11/dsh-ponytail
Lazy senior coding ruleset for DeepSeek Harness: injects minimal-code guidance, adds review, audit, debt, gain and help skills, and switches lite/full/ultra per session. Ships a script to sync rules from a pinned upstream commit.
dsh-cura
slow-stack/dsh-cura
A life-care layer for DSH: health log and cycle care in every session, saved coordinates so nearby-place questions do not need re-asking, Amap and Luckin MCP tools, plus a QQ companion layer that injects expression rules into replies.
dsh-slowmist-security-guard
wjl906/dsh-slowmist-security-guard
Security guard for DeepSeek Harness (Cordis): reviews every tool call in pre-execute against SlowMist agent-security rules, denying or flagging risky commands, MCP servers and skill loads, external URLs and repositories, untrusted documents, and EVM or Solana addresses; scans tool results for prompt injection after execution and ships a slowmist_scan audit tool plus an injected security policy prompt. Rules follow the open-source SlowMist Agent Security Skill (MIT).
dsh-thinking-channel-guard
laoyingfuhuole/dsh-thinking-channel-guard
Moves reasoning that a provider leaks into the visible text channel back into the reasoning channel, and emits a text block when a whole turn of answer text lands only in the reasoning channel. Catches three shapes: `<thinking>`-tagged text, text that continues a reasoning tail which ended mid-token, and an answer-only reasoning turn that stops. Healthy streams pass through unchanged, block order is preserved, and each of the four rules can be switched off in one JSON config.
dsh-plugin-antigravity-rules
domush/dsh-plugin-antigravity-rules
Loads Google Antigravity rules from ~/.agents/rules and the workspace's .agents/rules into the system prompt: always-on rules in full, the rest as a catalog the model can load with the antigravity_rule tool, or that activates when a touched file matches a rule's globs. The rules are prompt guidance rather than hard enforcement, and the model can toggle them off with the same tool. Settings are written to .agents/rules/settings.json in the workspace.
dsh-memory-nebula
hanhan9999999999999999999/dsh-memory-nebula
Local-first memory nebula for DeepSeek Harness with graph views, evidence chains, hygiene checks, and offline snapshots. Reads the memory stores of every workspace and of other memory plugins (read-only, for example meow-memory's SQLite) and writes only under $DSH_HOME/memory-nebula. Registers 9 agent tools and a system-prompt section (usage rules every turn, plus not-yet-read neighbouring memories after a few reads); its /export and /share endpoints return the full memory text without authentication on the local server, and an optional standalone engine listens on 127.0.0.1:7890.
dsh-plugins (archive-rsi)
feahter/dsh-plugins
Session archive with tags, notes and per-session metrics, plus a Plan-Do-Check-Act loop over the archived sessions that produces prompt and AGENTS.md rules and re-measures them on later sessions. Runs locally without calling a model.
dsh-session-icon
masterltx/dsh-session-icon
Draws an icon in front of every session row in the DSH sidebar, decided from the session itself: keyword rules on the title first, then the session content (title, first user message and tool names), then an optional one-shot model verdict, then a workspace fallback. 445 Ant Design outlined icons are inlined for offline use, and the rules, colours, keywords and row indent are editable in a settings page.
dsh-guard
netcatty/dsh-guard
allow/ask/deny permission gates for DeepSeek Harness tools: path/params/agents surfaces, argv decomposition, network rules, redaction, grants, dry-run, fail-closed config
dsh-liars-dice
dingchenhui0618-arch/dsh-liars-dice
A hot-seat Chinese Liar's Dice table for DeepSeek Harness: 2–6 seats, 3/5/7 dice, wild ones, an LLM opponent that only sees its own hand and is re-validated by the rules, a post-reveal referee, a replay that keeps the current round's dice hidden until the challenge, and a switchable exact-binomial probability panel. 54 node --test tests, zero runtime dependencies, four real rendered screenshots.
dsh-safety-inspector
ljttt0824/dsh-safety-inspector
Read-only behavior audit for DSH sessions: flags unsafe actions in 8 categories (A–H) with rules for 14 subclasses (the 41-subclass taxonomy file is not shipped, so a built-in 14-subclass fallback is used), anchors every finding to a real event seq, and gives deep links plus the full tool-call / data-access chain of any assistant reply (parameters, results, files read and written, domains, sandbox and approval context). Every finding is drill-downable.
mcp-hub
jiale-cheng-ning/mcp-hub
DeepSeek Harness tools to discover MCP configurations across clients, audit static security rules and probe a named server, using a separately installed mcp-hub Rust binary.
dsh-industry-tax
feely0208/dsh-industry-tax
Keyword-rule scan of a project directory for tax-industry compliance gaps ahead of regulatory audits: accounting voucher and ledger retention set under 30 years, invoice-title lines with no taxpayer ID, tax data leaving China with no security assessment or standard contract mentioned, and e-invoice lines with no verification or de-duplication mentioned, with rule sources kept in rules/.
dsh-ps-preflight
dingchenhui0618-arch/dsh-ps-preflight
Grades every PowerShell command an agent is about to run before it runs: deletion, overwrite, network download and credential access are four required classes, plus dynamic execution, system and privilege changes. Verdicts are a pure deterministic rule function — no model call, no network, sub-millisecond — with two escalation rules (download+execute, credential+exfiltrate) that jump straight to high. High risk stops the tool call for a second confirmation (in-panel queue or the native approval), deny or timeout means it never runs; medium is logged but never blocked. Every command and its verdict, including the redacted command text, is appended to a local JSONL audit log. Quoted danger strings are treated as data, -WhatIf is respected, -EncodedCommand is decoded and analysed again. Host + client halves, zero runtime dependencies, 55 rules, 136 tests.
dsh-skill-audit
vamyane/dsh-skill-audit
Skill and plugin audit console for DeepSeek Harness. It scans local skill folders and installed DSH plugins, checks the packaging rules that commonly break installation, scores each asset, and exports a Markdown fix report.