Плагины
Находите, фильтруйте и устанавливайте плагины DeepSeek-Harness.
Найдено плагинов: 225
dsh_use_browser (use-browser)
ankye/dsh_use_browser
Универсальная автоматизация браузера с шестью инструментами (browser_open, browser_click, browser_type, browser_extract, browser_screenshot, browser_eval), ссылками на элементы @e и двумя бэкендами: Playwright-headless / CDP (подключение к вашему собственному Chrome).
dsh-langfuse-plus
flysnaily/dsh-langfuse-plus
Наблюдаемость через Langfuse для DeepSeek Harness (DSH): деревья трассировок по каждому ходу, агрегированные по идентификатору сессии, с версионированием промптов и оценкой.
dsh-replay-lab
tbxy09/dsh-replay-lab
Повтор реальных ходов DeepSeek Harness против Standard, Minimal, Anchored или кандидатов-плагинов с зафиксированными доказательствами поверхности запроса.
dsh-trapstreet
trapstreet/dsh-trapstreet
Проверьте, какие плагины DeepSeek Harness действительно загрузились, и посмотрите публичные доски оценок на trapstreet.run.
dknowc-dsh
dylanzhangzx/dknowc-dsh
Доверенный офисный набор для государственной/корпоративной работы от Beijing Caizhi Technology (深知可信智能): три навыка — доверенные консультации (вопросы по политике/нормативам с маркерами цитирования + HTML происхождения), доверенный поиск (поиск авторитетных политик/стандартов с прослеживаемым HTML + чистым Markdown) и написание официальных документов (черновик/переписывание/Word/поставка на бланке); все интерфейсы идут через MCP доверенной среды dknowc (Bearer-авторизация), результаты попадают в рабочее пространство сессии, настройка ключа сохраняется для будущих сессий; публикуется на npm как dknowc-dsh.
dsh-browser
ben7am1n/dsh-browser
Автоматизация браузера для DeepSeek Harness: инструменты open/click/type/screenshot/eval на базе Playwright, позволяющие агенту управлять реальными веб-страницами.
design-playbook
bandersnatch0x/design-playbook
Design I/O для кодинг-агентов: управляемая генерация UI через декларации (spec/domain/craft/design/components/template) и контракты (skill/evaluator). Для продуктового UI — консоли, дашборды, agent-ops, приложения с приоритетом CJK.
dsh-context-proxy
evilirving/dsh-context-proxy
Лёгкое извлечение контекста по требованию: инструменты context_query / context_slice / context_grep, читающие уже сохранённую историю с безопасными для повтора цитатами.
dsh-py-runner
amzingluo/dsh-py-runner
VSCode-style Python console in the DSH sidebar: it opens .py files in an editor with line numbers and Python syntax highlighting, lets you pick the interpreter, and runs the file with Run / Debug / Stop while stdout, stderr and the exit code stream into a panel below. Debugging drives a real DAP session through debugpy: click the gutter to set breakpoints, then step, pause, inspect the call stack, expand variables and evaluate watch expressions; a missing debugpy can be installed from the panel in one click. Ctrl+Enter runs, Ctrl+Shift+Enter starts debugging. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-memory-es
xqh0405/dsh-memory-es
DSH memory plugin backed by Elasticsearch: per-user indices with index-level ACL, a shared team knowledge base, Chinese BM25 plus vector retrieval, and a built-in review and audit panel. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
feiyu-memgate (dsh-feiyu-memgate)
n0rmal-114/feiyu-memgate
Long-term memory through the bundled feiyu-memgate gateway, which the plugin starts on 127.0.0.1 (needs Python 3.11 or newer, nothing to pip install): recalls related memories at the start of each user turn and appends them after the user message as a marked data block, gives the model memgate_remember and memgate_search tools, and adds a /memgate command and a settings page where each feature can be switched off. By default memories are written only when the model calls the remember tool. Keyword retrieval over SQLite by default, with no model and no download; an OpenAI-compatible embeddings endpoint or the optional dsh-feiyu-memgate-embed add-on (local CPU embedding models, downloaded on request) adds semantic retrieval. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-weknora-memory
rennimars/dsh-weknora-memory
Layered context engine for DeepSeek Harness: per-kb WeKnora retrieval fused by RRF, recency decay and hot/warm/cold tiering, standing profile cards, fact-conflict surfacing, plus the four WeKnora tools (list / search / read / ask) for pull-on-demand retri
dsh-superagent-trace
luckyxiang6688/dsh-superagent-trace
Read-only explorer for what the Super-agent did with each message: the sources page lists inbound events per source with where they went (admitted, decided into a ticket, filtered, or dropped at the source), and opening a ticket draws its Router/Worker swimlanes, evaluation, dispatch, tool calls and replies on a timeline. Reads orchestrator.sqlite and session logs; writes nothing.
dsh-plugin-model-eval
zhangzhengruinus/dsh-plugin-model-eval
Evaluate your available models, compare capability and task time, and choose a configuration that fits your work.
dsh-research-solver
loneleave/dsh-research-solver
Research-style problem solving in four gated stages: record sources by retrieval tier with local before web, turn each shortfall of existing tools into a gap that needs evidence, score candidates against weighted metrics with a deterministic engine, fuse or improve the weak ones, then iterate on measured deltas. It first indexes the whole machine in two layers (what it can run — runtimes, toolchains, package libraries — and what it holds — code, docs, images, video, models), then organizes a directory into human-facing and machine-facing trees.
dsh-browser-ops
dcd887/dsh-browser-ops
Self-managed persistent browser for DSH automation — launches its own Chrome/Edge profile, navigates, switches tabs, snapshots with semantic refs, acts by ref, evaluates JS, screenshots, and uploads files. Does not bypass CAPTCHAs by design.
dsh-Engram
yhprime/dsh-engram
Experience and skill engine with a sidebar console and exp_record / exp_recall / exp_feedback / exp_consolidate / exp_type tools, stored in a local JSON file. Optional Git backup (off until you configure a remote) pushes the library to a remote you choose; GitHub sign-in uses a device code or a token kept in local settings.
socratic-tutor
icefox12138/socratic-tutor
Socratic tutor tools for DSH: cited retrieval, page images, chapter index, session review, embedded console. Teaching behaviour lives in dsh-socratic-persona.
dsh-drift
daphne999/dsh-drift
Capability drift detection: audit a plugin once, store a baseline of what it could do, then after every upgrade report only what is NEW (new install scripts, shell/eval access, outbound domains, credential paths). Deletions and version bumps are not reported, and no score is computed.
dsh-daughter
han0han0le0ni/dsh-daughter
Five-layer persona (L1–L5), event memory with FTS5 plus Dice retrieval, cognitive filter, skill cards with auto-frozen asset packs, millisecond reflex rules, three-tier safety degradation, a read-only brain star map, and an optional VPet-Simulator MOD bridge.
dsh-elysia-memory
1148462398-hub/dsh-elysia-memory
Elysia's memory plugin for dsh. One shared, human-editable MEMORY.md across every session: two-disk mirror, history snapshots, cross-process write lock; entries tier between the main file and a searchable archive by hit frequency; retrieval is keyword co-occurrence plus BM25. Nine sections (persona, requirements, corrections, lessons, methods, requests, sessions, method library, ingested docs) plus a dedicated self-identity section for a character's own self-cognition, injected above the persona. Ships an opt-in memory pack (packs/elysia: 61,903-char identity + 2,482-char persona, distilled from 1,696 officially-confirmed voice lines and 112,175 dialogue lines) that is off by default, backs up before writing, only touches identity/persona, and can be uninstalled. Also opt-in voice playback through a local GPT-SoVITS engine, a skills library, and a sticker library that sends a fitting sticker at the right moment. On by default: after every reply a bubble with an avatar pops up (voice.bubble: false turns it off), and unless the agent already spoke itself it costs one extra small-model request. Opt-in voice playback needs a local GPT-SoVITS engine (about 27 GB, installed separately, NVIDIA GPU required; the plugin downloads about 200 MB of weights) and is Windows-only; the automatic second-drive mirror is Windows-only too (elsewhere set mirrorPath, otherwise there is no mirror).
dsh-anic
xuanyuanchumo/dsh-anic
AI-native iteration cycle: 24 anic_* tools and 77 HTTP routes that run evaluation, convergence gates, and auto-fix loops across projects, with a live white-box panel (interactive workflow graph, invocation feed, autonomy gauge, gate ledger, HITL queue) and an inline installable SKILL.md workflow kit.
dsh-Immunity-memory (memory-plugin)
thebeautifulice/dsh-immunity-memory
Cross-session long-term memory for DSH: eight memory layers, four notebooks, hybrid SQLite FTS5/vector/graph retrieval, agent-driven automatic capture, write lifecycle hooks, optional write audit, project spaces, forgetting and a visual management page. Full FTS5/vector/graph retrieval needs the bundled Go service, which ships only as a Windows x64 prebuilt; on macOS/Linux, build it from source or the plugin falls back to Node SQLite without FTS5 and retrieval degrades.
dsh-elysia
1148462398-hub/dsh-elysia
Elysia's memory plugin for dsh. One shared, human-editable MEMORY.md across every session: two-disk mirror, history snapshots, cross-process write lock; entries tier between the main file and a searchable archive by hit frequency; retrieval is keyword co-occurrence plus BM25. Nine sections (persona, requirements, corrections, lessons, methods, requests, sessions, method library, ingested docs) plus a dedicated self-identity section for a character's own self-cognition, injected above the persona. Ships an opt-in memory pack (packs/elysia: 61,903-char identity + 2,482-char persona, distilled from 1,696 officially-confirmed voice lines and 112,175 dialogue lines — not written by hand, measured) that is off by default, backs up before writing, only touches identity/persona, and can be uninstalled. Also opt-in voice playback through a local GPT-SoVITS engine, a skills library, and a sticker library that sends a fitting sticker at the right moment.