DeepSeek Harness Plugin Hub

发布与管理完整 Harness Profiles,发现适合你的插件。

探索

插件目录环境预设文档中心动态

社区

发布插件联系我们报告问题

相关链接

Plugin Hub GitHubDeepSeek Harness 官方项目系统状态隐私说明
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

独立、非官方社区项目,与 DeepSeek 官方无隶属、授权或背书关系。

Mimo Agent Tools — DeepSeek Harness 插件(DSH Plugin)
DeepSeek Harness Plugin Hub
ProfilesPlugins分类动态文档登录管理 Profiles
ProfilesPlugins分类动态文档登录
← Plugins
M

dsh-mimo-agent-tools

Mimo Agent Tools

面向 DSH 智能体的 Xiaomi MiMo 搜索与多模态工具:mimo_search/vision/audio/video/asr/tts。

插件会安装到这里;不确定时保持 web。

npx -y @deepseek-ai/dsh plugin --profile web add github:ch1bug/dsh-mimo-agent-tools#ceb833b9059d4a2d220c180f0b72db6734c0e62b
README兼容性版本

兼容性与来源证明

Mimo Agent Tools 以 dsh-mimo-agent-tools 发布,当前版本为 0.2.1-alpha.1-r1。Plugin Hub 会校验它的 manifest,并保存精确安装来源,便于复现安装结果。

DSH 兼容范围
*
运行环境
any
发布来源
github
Registry 更新时间
2026/10/3

版本

0.2.1-alpha.1-r1prerelease
2026/10/3
0.1.0stable
2026/9/30

相关插件

正在加载相关插件…

最新版
0.2.1-alpha.1-r1
DSH
*
HMR
重启进程
Tree shaking
未声明可安全裁剪
解包体积
未提供
文件数
未提供
Surface
any
许可证
MIT
发布源
github
GitHub
★ 0
周下载
0
查看源码 ↗
README Badge

点击下方 Badge 复制 Markdown,粘贴到 README 即可。

这是你的 Plugin?认领权益 · 优先安全扫描

验证 package.json 声明的 GitHub 仓库,即可管理这个公开页面。认领后,Hub 会优先安排当前版本的安全扫描,并在通过后公开展示结果。

认领这个 Plugin →
报告问题

相关插件

继续浏览 vision-media 分类下经过校验的插件。

Modlens@liustack/modlens面向仅支持文本的 LLM 的插件视觉能力,由免费的 Antigravity CLI 提供支持Tool Describe Image@linxin666/dsh-tool-describe-image面向模型的 describe_image 工具,用于 dsh Web GUI:通过在兼容 OpenAI 的端点调用视觉语言模型,为文本模型提供图像理解能力,以描述一张图像(本地路径、http(s) URL 或附件引用)。可热插拔 —Vision Routerdsh-vision-router为仅支持文本的 DeepSeek Harness agents 提供视觉能力:内置免费视觉链路和像素级工具。除非配置为严格仅限本地使用,否则远程视觉提供商(包括默认的免费回退服务)会接收图像内容。Vision Toolkit@anionex/dsh-vision-toolkit面向 Harness 原生集成的 DeepSeek 与 agent-vision-toolkit:图像问答、OCR、定位、UI 还原、像素差异、Artifacts 和 Web UI。

README

mimo-agent-tools

DSH (DeepSeek Harness) Cordis plugin that turns the Xiaomi MiMo API into model tools for an agent: web search, image/audio/video understanding, speech-to-text, and text-to-speech.

Backed by the OpenAI-compatible endpoint https://api.xiaomimimo.com/v1.

Tools

ToolModelPurpose
mimo_searchmimo-v2.5-proWeb search via the native web_search tool, returns answer + cited sources (incl. site_name/logo_url); optional force_search, user_location
mimo_thinkmimo-v2.5-pro / mimo-v2.5Deep thinking — full reasoning chain (reasoning_content) + final answer
mimo_jsonmimo-v2.5-pro / mimo-v2.5Structured JSON output via response_format: json_object
mimo_visionmimo-v2.5Image understanding — local files (auto base64) or public URLs, multi-image
mimo_audiomimo-v2.5Audio understanding / transcription (wav/mp3/flac/ogg/m4a) — local file or public URL
mimo_videomimo-v2.5Video understanding (mp4/webm/mov) — local file or public URL; optional fps / media_resolution
mimo_asrmimo-v2.5-asrSpeech-to-text with optional language hint — local file or public URL
mimo_ttsmimo-v2.5-tts / -voicedesignText-to-speech to a .wav/.mp3 file — preset voices or free-form voice design; optional style (speaking tone) and format (wav/mp3)
mimo_voiceclonemimo-v2.5-tts-voicecloneVoice cloning — reference clip + text → speech in that voice; optional format (wav/mp3)

audio-tools skill

The plugin registers an audio-tools skill that guides when to use the four audio tools (mimo_asr, mimo_tts, mimo_voiceclone, mimo_audio). The tools themselves are always registered — they are lightweight pure-API calls — so the skill only teaches usage, it does not gate the tools.

Requirements

  • A MiMo API key — the web_search tool requires the connected-search plugin enabled in the MiMo console.
  • python3 on the host (used by driver/mimo_driver.py).
  • DSH host with the shell and sandboxPolicy services.

Configuration

The API key is resolved at tool-call time from the DSH credentials service first (key name XIAOMI_API_KEY — the web Models page writes keys there), falling back to the environment. Nothing is hardcoded:

SourceKeyPriority
DSH credentials service (~/.dsh/.credentials.yaml, web Models page)XIAOMI_API_KEY1
EnvironmentXIAOMI_API_KEY or MIMO_API_KEY2

Other options (environment, read at apply time):

Env varPurposeDefault
MIMO_DRIVERpath to driver/mimo_driver.py~/.local/lib/mimo-agent-tools/driver/mimo_driver.py
MIMO_TMPtemp dir for spec/response files/tmp

Install the driver at the default path (or point MIMO_DRIVER at it):

mkdir -p ~/.local/lib/mimo-agent-tools/driver
cp driver/mimo_driver.py ~/.local/lib/mimo-agent-tools/driver/

Why a python driver?

Multimodal payloads are multi-megabyte base64 strings. Agent shells commonly cap captured stdout (DSH's bash seam caps at 64KB), which silently truncates large payloads. The driver therefore does ALL file reading, body assembly and HTTP POSTing inside one python3 process, using spec/response files on disk — nothing large ever crosses the shell.

Install (DSH bundle)

Standard DSH bundle — install with the official plugin command (auto-inits the profile, pnpm-installs, and appends the bundle layer per dsh.bundle.patch):

# From a local checkout, or via git/npm:
dsh plugin --profile web add /path/to/dsh-mimo-agent-tools
# or: dsh plugin --profile web add github:you/dsh-mimo-agent-tools

# Install the python driver to the default path (MIMO_DRIVER points at it):
mkdir -p ~/.local/lib/mimo-agent-tools/driver
cp driver/mimo_driver.py ~/.local/lib/mimo-agent-tools/driver/

# Restart dsh web; the tools mount automatically.

Dependencies are declared as peerDependencies (ecosystem convention — @deepseek-ai/dsh-tools is already loaded in the DSH process, so nothing is duplicated). dsh plugin add installs the bundle into the profile's node_modules where peer deps resolve against the running harness.

Notes on the MiMo API (from the official docs)

  • TTS target text goes in the assistant message; the voice description (voicedesign model) goes in the user message; a style instruction rides the user message for preset voices and becomes an inline (风格) tag prefix for voicedesign voices.
  • mimo-v2.5-tts-voicedesign does not accept an audio.voice field — it uses optimize_text_preview instead.
  • ASR (mimo-v2.5-asr) must not receive a thinking field.
  • input_audio.data / video_url.url accept either a public URL or a data:<mime>;base64,... data URL (video base64 capped at 50MB per the docs).
  • Web search costs per keyword round (max_keyword, default 3) — see MiMo pricing. force_search (default true) trades freshness against cost; user_location biases results, e.g. {"type":"approximate","country":"China","region":"Hubei","city":"Wuhan"}.
  • Deep thinking (mimo_think) returns reasoning_content + content; in multi-turn agent conversations with tool calls, reasoning_content from earlier turns must be echoed back or the API returns 400.
  • Structured output (mimo_json) needs an explicit JSON shape description in the prompt (fields, types, nesting); keep max_completion_tokens generous so the JSON is not truncated mid-document.

Tests

python3 tests/test_driver.py      # driver request-body assembly (14 cases)
node --test tests/tools.test.mjs  # tool registration surface (10 cases)
node --test tests/speech.test.mjs # speechToText provider surface (4 cases)

Official voice input (speechToText provider)

speech.js is a separate plugin entry that plugs Xiaomi MiMo ASR (mimo-v2.5-asr) into dsh's official voice-input stack as a cloud provider: the built-in 🎤 capture → RPC → provider-registry pipeline stays untouched; recordings are transcribed by MiMo and the transcript lands in the input box.

  • The entry stays dormant (inject: ['speechToText', ...]) on deployments without @deepseek-ai/dsh-experimental-voice-input-bundle — the agent tools are unaffected either way.
  • The bundle patch overrides the speech-to-text row's defaultProvider to mimo. Bundle order matters: this package must compose AFTER the voice-input bundle in dsh.profile.bundles (the override only lands on an already-inserted row). SenseVoice stays registered and can be re-selected in the official voice settings.
  • Credentials: the same XIAOMI_API_KEY entry in the DSH credentials service the tools use.

License

MIT