Plugins
Browse, filter, and install DeepSeek-Harness plugins.
288 plugins found
dsh-review-loop
wuxiangru915/dsh-review-loop
Incremental diff reviewer for DSH: after approving a batch, only new changes since the checkpoint are shown, with a Web review panel and `/review` command that inject feedback back to the agent.
dsh-git-guard
bibibala/dsh-git-guard
Git-aware write guard plugin for DeepSeek Harness: blocks whole-file writes that would overwrite the user's uncommitted changes, and reports what happened to protected files at turn end
dsh-git-panel
lcx19970714/dsh-git-panel
Git panel for viewing diffs, editing files, staging and committing changes, and generating commit messages using models configured in DSH.
dsh-whale-glance
yan9690/dsh-whale-glance
Screen companion for DeepSeek Harness on Windows. A background helper samples the screen, diffs it, and when the picture changes enough, wakes a one-shot subagent that looks at a stitched frame grid plus the on-screen text, and writes one line to a floating subtitle window. Game mode recognizes the running game and uses its slang library; video mode reads subtitles so no line is missed and can keep a per-show episode archive. The subagent never writes back into your main session, so long use does not grow its context. Off by default, with cooldown, hourly cap and window-title filter.
dsh-native-tools
brightstonebody/dsh-native-tools
Four native macOS tools in one plugin — drag-to-screenshot with annotation and OCR, clipboard history, select-to-translate, and a menu bar system monitor (network up/down, CPU and memory, each clickable for per-process rankings). All four features share a single AppKit helper process, are toggled independently, and apply setting changes without restarting the plugin or the process.
dsh-knowledge
avenyq/dsh-knowledge
Sidebar knowledge base for DSH: indexes local Obsidian vaults and project folders into SQLite FTS5, re-indexes file changes automatically, shows source snippets while you type, and lets the agent answer with numbered citations that open the exact source file and line. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-journal
isdou/dsh-journal
Generate structured work reports and summaries from DSH conversations and file changes, with local history and Markdown export.
dsh-document-view
frank-fan-1058/dsh-document-view
A DeepSeek Harness bundle that shows an opened document in the center while the current conversation stays visible on the right, with a per-session toggle, a resizable divider, layout presets and a narrow-window Document/Chat switch. Only the window layout changes; editing, saving and preview continue to use the host's native document and Office viewers.
dsh-codex
zhang6hk/dsh-codex
Review changes or delegate investigation and fix tasks to the local Codex CLI.
dsh-app-icon
jiaxusummer/dsh-app-icon
Changes the DeepSeek Harness desktop app icon (window, taskbar, tray): use your own image or restore the official set. DSH quits and restarts once to apply it. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-tool-keeper
wufeng0105/dsh-tool-keeper
AI re-reads every tool's manual on every reply, including the tools you never use. Hide the ones you don't need and rewrite the manuals of the ones you keep — a tool's own manual or any single parameter's. Rewriting touches only the wording, never the schema, so a rewritten tool still works. Both the settings page and a model-callable tool write the same config; changes apply after a restart. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-session-ref-unlimit
imcaibaicai/dsh-session-ref-unlimit
Lifts the built-in cap of three @ referenced sessions per message by rewriting the reference resolver's config after it is instantiated, defaulting to 8 with an aggregate context budget guard; no app.asar changes.
dsh-model-endpoint-link
imcaibaicai/dsh-model-endpoint-link
Broadcasts the current session's effective model selection as a window event at the moment it changes, for consumers that need the selection before the first request round, such as a balance widget switching its accounting endpoint.
dsh-model-sync-plugin
chen1pengvincent/dsh-model-sync-plugin
Discovers the model providers actually configured in DSH, surfaces unread catalog changes from eight official platforms, and applies only the selection you confirm, with a version check and a pre-write backup.
dsh-cool-plugins (dsh-cool-git)
cooltea001/dsh-cool-plugins
Git plugin — adds a Git panel to the session tab bar: changes, commits, history, branches, remote sync, and merge-conflict resolution in one place, with AI-written commit messages.
dsh-rightbar-max-keep-rail
nohappiness/dsh-rightbar-max-keep-rail
Changes the right sidebar panel-fullscreen mode to cover the conversation only: the panel starts at the left rail right edge instead of spanning the whole window, so the session list stays visible while the panel is maximized, and it keeps following the rail as you widen, collapse or drawer it.
dsh-web-control
weichengwu/dsh-web-control
Manage a LAN or Tailscale reachable dsh web instance from inside DSH. dsh web listens on loopback by default; this adds a Settings > Web access panel that detects the machine's Tailscale and LAN addresses, starts dsh web bound to the one you pick, and shows the token URL to open on a phone or another machine. Also stops and restarts it, tails the child log, persists the bind address and port, and registers two agent tools (dsh_web_control, dsh_web_config). Zero runtime dependencies. Anyone who can reach the bound address and holds the token URL gets your dsh's full privileges (files and commands); LAN mode is plain HTTP, Tailscale is encrypted by the tunnel, and the token is single-use and changes on every start.
dsh-team-toolkit (team-ext)
dreamperday/dsh-team-toolkit
Observation-only audit for DSH Agent Teams that records per-request cache usage and prompt-prefix rewrites to JSONL, with no tool or prompt changes.
dsh-qa-isolation
shadowtang98/dsh-qa-isolation
Adds a Q&A agent preset that answers each new question from a clean context: earlier exchanges leave the model input via a surface replacement, with no extra model call and no summary residue. A /qa follow command and a composer toggle switch to follow-up mode, and /clear archives the session and opens a fresh one under the same preset.
dsh-session-fold
kennethfan/dsh-session-fold
Folds older oversized tool results into spill files and replaces them with short, deterministic stubs at step boundaries, with no LLM call: the stub keeps the first 200 and last 80 characters plus the sequence number, tool name, line/char/size counts and the original file path, and each fold is a real surface replacement accompanied by compaction/prune shadow pricing so token counts stay honest. Triggers on context pressure at 25% of the model window (falling back to 32768 tokens when the window cannot be resolved) or on five accumulated candidates; the last three turns, error results and rich content are never folded. The original is recovered from the session log rather than the spill file, so /folds, /unfold <seq> and /unfold all keep working after spill cleanup; session_fold_recall lets the model fetch a folded result by sequence number or by keyword, and session_fold_diagnose reports the effective settings. /spill-prune removes orphaned spill directories, preview only unless --apply. A Settings page exposes 12 of the 13 options as draft-then-save fields; they are volatile, so saved changes apply without restarting the plugin.
dsh-context-domains
guicaiyue/dsh-context-domains
Puts a domain bar under the session header — a domain groups skills and tools and toggles as one block, so one click loads or unloads every member. Hovering a chip lists its members (skills, then tools) in a height-capped panel that scrolls. The right end of the bar is a compaction-threshold counter (− 5 +, 100k tokens per step) that rewrites the live compaction engine policies, so the compaction trigger line changes on the next turn without restarting dsh; a degraded fallback turns the number amber and is reported in the snapshot. Settings maintains domains and members with search, skill/tool tabs, all/selected/unselected filters and a copyable AI grouping prompt. Tool removal uses tools.restrict plus a guard; hiding a skill re-registers it with modelInvocable false, so it leaves the model catalog while /name still works. Sessions of another preset are left untouched.
dsh-big-coming
sumcam-l/dsh-big-coming
Watches the official DeepSeek models for changes: every cycle it reads the model catalog and sends one minimal completion per watched model, then compares the system fingerprint, the served model name, the response envelope shape, the catalog metadata and the token count of a fixed prompt. When something moves, it pops a random meme with a table of exactly what changed and plays a blast. Needs a DeepSeek API key (kept in the credential store); by default it runs every 60 minutes and spends tokens on each watched model.
qq-integration
mkiea/qq-integration
Derivative of cheesehaqi/dsh-qq-onebot-bridge with the same package name, so the two cannot be installed together. Bridges a personal QQ account into DeepSeek Harness over OneBot 11: private and group messages become per-conversation agent sessions and replies are sent back to QQ. Ships a reverse WebSocket server, allowlists (empty = deny all), @ parsing, image and SILK voice handling, and a separate local console on 127.0.0.1:8880 with the decision trace, offline replay, event injection, a read-only preflight and host/Waylay start-stop controls. Needs a separate OneBot v11 implementation (a third-party unofficial QQ client): the default, Waylay, injects into the official Windows QQ.exe (Windows 10/11 x64 only) and is distributed as a separate release archive, not bundled in this repo. botQqAllowed and replayGuardEnabled are edited in DSH Settings > QQ助手, and the replay guard is armed automatically when the logged-in account changes.
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.