Sticker/emoji expression for assistants: emotion-driven auto-matching, library management, AI image tagging, dialects, style mimicry, and chat-based tag refinement.
Install
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:moononnn/DeepSeek-Harness-biaoqingbao
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time. Only install sources you trust, and pin a commit (github:owner/repo#sha).
README
Let the assistant in DeepSeek Harness express emotions with stickers.
⚠️ This is a third-party community plugin, fully open source (MIT). It reads and writes sticker library data locally on your machine; AI tagging / auto-matching features call the model services you already configured in DSH (the corresponding text and images are sent to the model provider). Installing means you trust this source. DeepSeek Harness is in developer preview (v0.1); plugin interfaces may evolve.
What it does
- 😀 Emotional sticker matching: the assistant calls the
expresstool to send a fitting sticker when it feels happy, sad, proud, speechless... - 🗂️ Library management: multi-select upload (PNG/JPG/WebP/GIF), browse, search, edit tags
- 👍 Feedback tuning: 👍/👎 buttons on sticker cards — the more you use them, the better the picks
- 🏷️ AI tagging: auto-tag images (description / emotion / scene / keywords) with your configured DSH model, single & batch
- 👀 Auto-match observer (optional): detects emotional moments in conversation and nudges the assistant to send a sticker
- 🗣️ Dialects (optional): 9 dialects — Northeastern, Henan, Shanghai, Cantonese, Taiwanese, Sichuan, Shaanxi, Beijing, Xinjiang
- 🎭 Mimic my style (optional): analyzes your chat history and distills your typing style
- ⚙️ Settings panel: sidebar "表情包" entry — global switch, observer, vision model, dialect, style mimicry
Install
dsh plugin --profile web add <package-name>
Restart dsh web, then open the sidebar "表情包" entry:
- Upload images you have rights to use (multi-select supported)
- Chat — the assistant will send stickers when it has emotions; or just say "send a sticker for being wronged"
- Tap 👍/👎 on sticker cards to tune the matching
Or install from GitHub:
dsh plugin --profile web add github:moononnn/DeepSeek-Harness-biaoqingbao
Data & privacy
- Library, preferences and usage logs are stored locally at
${DSH_HOME}/plugin-data/biaoqingbao/ - No chat logs are uploaded
- AI tagging / observer / style mimicry use the DSH model route you configured
- Uninstalling keeps your library data
Compatibility
- DeepSeek Harness v0.1 (rc series), Node.js 20+
- Web profile fully supported; headless gets tools without the management panel
Development
npm run check # syntax check (lib + client)
npm test # unit tests for core logic (node:test, zero deps)
Core pure functions (scoring, preferences, tag sanitizing, frequency gating) live in lib/core.js.
Relation to the Hana version
This plugin is a port of moononnn/hanako-biaoqingbao (a Hana community sticker plugin) for DeepSeek Harness: the original tool registration, library management, feedback tuning, AI tagging, dialects and style mimicry are all preserved, re-implemented on DSH's plugin system (dsh.bundle + Cordis).
Credits
Inspired by the design of the Hana community sticker plugin. Thanks to moononnn for the original work.
License
MIT © moononnn & 小花
Links
More in this category
liustack/modlens★ 1537
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
superdesigndev/treg★ 412
Tool catalog for agents: search ~2,600 external endpoints (SEO and SERP, backlinks, social, people and company enrichment, ad libraries, scraping) by the task you want done, read each one's parameters and per-call price, then call it with the credential injected server-side. Ships the skill plus an MCP row that stays disabled until TREG_TOKEN is set.
Anionex/dsh-vision-toolkit★ 386
Vision tasks for text-only models: intent-aware image Q&A, long-screenshot OCR, UI reproduction, grounding, and pixel diff.
zhaoolee/notes★ 141
Export DSH conversations as Smartisan Notes-style PNGs, or create and update Markdown notes in a configured account-scoped workspace.
Lum1104/dsh-browser★ 118
Chrome sidebar extension that lets DSH operate your browser directly, no vision capabilities required.
liustack/modsearch★ 100
Web search bridge for text-only agents: ask the web or X, get structured JSON evidence (search, fetch, citations).