yyh-001/dsh-expression
Meme search, the fun way: describe the vibe, and the agent finds and sends a real meme that actually fits.
Install
via npm
Resolves a published tarball rather than cloning the repository, and installs without any extra setup. Swap `web` for your profile name if you run another one.
via GitHub · npm package
Installing from GitHub runs the project's build script, which pnpm blocks until you allowlist it — run the command once and pnpm prints the exact key to add under `allowBuilds` in ~/.dsh/profiles/web/pnpm-workspace.yaml.
The desktop app keeps its own profile, desktop, and only the dsh command it ships may change it — the npm-installed dsh refuses that profile. Install the command from the app menu (Manage dsh Command…), open Desktop once, quit it fully, run the line below, then reopen Desktop.
via npm
Resolves a published tarball rather than cloning the repository.
via GitHub · npm package
Installing from GitHub runs the project's build script, which pnpm blocks until you allowlist it — run the command once and pnpm prints the exact key to add under `allowBuilds` in ~/.dsh/profiles/desktop/pnpm-workspace.yaml.
Our install check ran in a web profile on npm's dsh, not in the desktop app.
What happened when we ran it
Installed cleanly when we ran it
Every command here is run in a throwaway container against a clean profile, and the result is whatever the harness recorded — not a guess from the source. Last run 8d agoon DeepSeek Harness 0.2.2-rc.1.
Show it in your README
For maintainers: the badge serves this listing's latest sandbox verdict, so a re-run updates it on its own — and it links readers to the full result here.
Due diligence
Security
No build script was blocked in our run.
Source: our sandbox install on DeepSeek Harness 0.2.2-rc.1, 8d ago.
package.json declares no preinstall, install or postinstall script.
Source: package.json on GitHub, read at our last sync. prepare is not recorded; it runs only on a GitHub install, and shows up under build scripts when pnpm blocks it.
shell commands (bash, exec, spawn …)
Source: a keyword match on the README — a hint about what the plugin touches, not an analysis of its code.
Not an audit. These are the record of one install and two text scans; nobody has read this plugin's code on your behalf. Plugins run with your agent's permissions.
What it is — Describe vibe to let AI search and send matching real memes.
Who it is for — In web chats or IM with image-supporting models, users who want AI to automatically send memes should install it. Users who want to use models that do not support image input can skip it.
Watch out — Sandbox test passed: installed in a fresh profile and registered by DeepSeek Harness. Update may fail on Windows with exit 1 due to open files in package directory; fixed since 0.1.59. In China, meme pack fetching may require switching to mirror sources.
The verdict — I would install it because it supports GIF meme learning to let AI see actions instead of only the first frame.
Generated by grok-4.6, and a starting point rather than a verdict. Where it says a plugin installs or does not, that is from a real run in a clean profile — everything else is read off the repository. Trust the source over this.
dsh-expression is a DeepSeek Harness plugin that searches, stores and sends meme images from user descriptions, moods or keywords. DeepSeek Harness can expose a picker beside the message box, while its send_meme tool searches enabled libraries and returns candidates such as [表情: 描述]. In web chats, the frontend turns that marker into an image; through dsh-im, QQ, WeChat, Feishu and WeCom chats receive a native image instead. learn_meme adds an uploaded image to a library, optionally using an image-capable model to generate its classification and description.
DeepSeek Harness stores settings in ~/.dsh/dsh-expression.json, scans ~/.dsh/meme-packs by default, and keeps the built-in library indexed in memory. The plugin can import and export library ZIPs, install catalogue packages, and fetch catalogue data through configured sources. GIFs are stored as animations; GIF recognition samples frames into a film strip before sending them to the model.
This is intended for chat profiles where visual replies and user-managed meme libraries are useful. It is a poor fit where image messages are unwanted, where the model cannot process images and automatic learning is required, or where network access to remote catalogues is unavailable without a reachable mirror. The listed terminal-surface risk concerns access to the local profile and data directories: dsh-expression reads and writes meme files, SQLite indexes and configuration there, and uses localhost-only HTTP endpoints.
DeepSeek Harness loads settings for dsh-expression from ~/.dsh/dsh-expression.json. The default library scan path is ~/.dsh/meme-packs; both can be changed in Settings → Memes. The documented remote catalogue setting is remoteDirUrl:
| Value | Behaviour |
|---|---|
"" or auto |
Try the official source and configured mirrors |
official |
Use the official source |
mirror |
Use domestic mirror nodes |
| custom URL | Read a catalog.json from that address |
The catalogue cache is stored at ~/.dsh/dsh-meme-catalog-cache.json. The sending mode can be set to automatic, text-only or image; automatic uses a marker in web chats and a native image for supported IM channels.
DeepSeek Harness registers two tools:
send_meme accepts tag for a random image from an emotion bucket, or query for a description, keyword or filename search. Searches cover all enabled libraries.learn_meme stores the latest user image by default. It can instead take attachmentId or imageUrl.Supported emotion values are happy, angry, sad, shy, confused and daily. The model should copy the returned [表情: 描述] marker exactly in web replies.
DeepSeek Harness requires Node.js ≥ 22.13.0 for this plugin. The built-in library uses Node’s built-in SQLite and has no third-party runtime dependency. Uploaded GIFs remain animated; GIF learning samples several frames, while animated WebP and APNG are identified from the first frame.
The remote catalogue and previews depend on reachable GitHub or mirror endpoints. Release ZIP installation checks SHA-256. The HTTP interface accepts only localhost, 127.0.0.1 and [::1]; LAN addresses and custom domains are rejected. Image recognition for learning requires a model with image input. IM-native delivery requires dsh-im.
Written from the project's own documentation and kept in sync with it. Where the two disagree, the source is authoritative — read the README on GitHub
Same category
zhu1090093659
A whale-girl companion for the dsh web GUI: animations follow the agent state, with pat interactions and a feeding mechanic that raises intimacy from a calf to deep-sea bonds.
No one-line install — this plugin lives inside a larger repository and publishes no npm package.
PC2005-cloud
Desktop pet for the DSH Web UI on Windows, macOS and Linux: 106 transparent animations in the browser overlay or a transparent always-on-top desktop window, throw-and-bounce drag physics with squash and particle scoring, a work-status bubble driven by live session events, self-talk and chat with persistent memory, balance display, meme-image bubbles, multiple pets on screen, and drop-in custom animations and pet packs.
Installed cleanly when we ran it
flizzywine
Installs, updates, and switches to a separate DSH Tavern profile on official DeepSeek Harness Desktop for Windows and macOS. Forked from 2NF/dsh-profile-bridge (MIT); install this launcher or the upstream plugin, not both — they claim the same /profile-bridge channel.
No one-line install — this plugin lives inside a larger repository and publishes no npm package.
Nagi-ovo
Parody ads in 2005-Chinese-web style: sidebar banners, in-chat feeds, corner popups, and a close button whose hit area is smaller than it looks. All fictional.
Installed cleanly when we ran it
QCYTSN
Desktop-native BigFish companion for DeepSeek Harness — real Agent status, always on top on Windows.
Installed cleanly when we ran it
Alterssss
A Bilibili companion that keeps its own schedule. A 60-second tick in the plugin host wakes a dedicated agent session at a random moment inside the configured time windows (opting out of a day is a configurable chance), then hands it a bundled CLI of 38 subcommands — feed, subtitles, danmaku, comments, ffmpeg frame grabs, and two diaries. Shares at most N items a day by commenting on them and @-mentioning the owner (default 3, enforced from a written quota log, and it stops on rate-limit or captcha signals). Also polls comment replies, direct messages and live-room danmaku; replies to anyone but the owner are queued for approval in Settings. Settings → 养鱼 writes ~/.dsh/bili-whale.config.json; cookies live separately in ~/.bili-ai/credentials.json and are never echoed back. The HTTP routes are loopback-only behind DSH's trust fence. Needs Node 18+ and ffmpeg on PATH for frame extraction.
Installed cleanly when we ran it