modlens

DeepSeek's flagship chat models, and GLM-5.3 itself, are text-only and cannot read images. GLM-5.3-Flash is native multimodal. ModLens is a plug-in vision engine that gives a text-only model sight. ModLens reads images pasted straight into the chat, no saving to a file and passing a path first.

Talk to us

Issues are welcome any time: open one. Follow the liustack WeChat official account, and come find me on X: @liustack. What you built with it, which harness you are on, and what should come next are all shared on WeChat and X. A proper community space is on the way.

Highlights

🥇 The most capable vision plugin for DeepSeek Harness (dsh): install it instantly with one command: npx -y @deepseek-ai/dsh plugin --profile web add @liustack/modlens@3.27.2. See the setup guide for installation and update details. If the command line is not your thing but you still want to try DSH, check out AIManager, the lightest desktop wrapper for DeepSeek Harness. It gets you started with zero code or configuration and installs every dependency for you with one click.

Pasting an image works two ways. ① Just paste. On a text-only model the pasted image lands as a private temp file and its path enters the composer (the same interaction OpenCode and Pi ship), then the modlens_read_image tool takes it from there. ② Pick a (modlens vision) entry in the model selector (it remembers your choice, so once is enough), then paste: the thumbnail stays visible in your message, closer to the Codex app feel, and the image is converted to structured evidence at request time, answered by the same underlying route. The plugin auto-discovers every provider route carrying eligible text-only DeepSeek, GLM, or MiMo Pro models and adds a wrapped entry per route. A stock install gets DeepSeek-V4-Flash (modlens vision) and DeepSeek-V4-Pro (modlens vision), while extra routes like opencode-go or zai get their own. Native vision models in those families, including GLM-5.3-Flash, are excluded automatically. Which paste route applies is the host's per-model call: only a model its metadata positively confirms text-only is taken over, anything unconfirmed is left alone, so vision models keep their native paste (details).

Paste images directly in every harness. No saving to a file and passing a path first.

A hotkey that captures the screen into DeepSeek Harness is a separate plugin: dsh-screenshot.

• The lightest touch on the market. No hooks, no wrappers, no local proxy daemon, not a single line changed in any harness config: on the skill harnesses it is exactly one skill folder, on dsh exactly one plugin. Uninstalling is deleting a folder, and your agents are back to stock.
• Zero-config start. Reuses existing setup in Claude Code, Codex, OpenCode, and Pi, plus other multimodal models already on your machine. Nothing installed locally? Antigravity CLI is a free no-key channel, and a free Gemini key brings a read down to 5-10 seconds. API keys from every major OpenAI-compatible provider work too.
• Comma-separated keys rotate on auth, rate-limit, or quota failures. Other failures skip remaining keys and keep the existing provider failover.
• Evidence, not imagination. Full transcription, reading-order layout regions, entity and relation lists. The model quotes specifics.
• Install once, use everywhere. Verified on real machines in Claude Code, Codex, Pi, and OpenCode.

Install in other harnesses

Option 1, install with skills.sh:

npx -y skills add liustack/modlens --skill modlens --global

This installs the modlens skill at user level. Restart the harness, then ask your AI to configure modlens and run its health check.

Option 2, hand the install to your AI. Send it this line:

Install and configure the modlens skill following https://github.com/liustack/modlens/blob/main/INSTALL.md, then run the health check and tell me the result.

The install starts by checking what your machine already has. An existing login in Claude Code, Codex, OpenCode, or Pi can be enough: modlens asks before reusing any of them, and the health check tells you where things stand.

After either option, only if the health check comes back empty, set up a free engine. The recommended choice is a free Gemini API key (about three minutes at Google AI Studio, no credit card), which also makes every read 5-10 seconds. A free OpenAI-compatible key from another platform works too. To avoid any sign-up, install Antigravity CLI instead, then sign in:

curl -fsSL https://antigravity.google/cli/install.sh | bash
agy                                                           # sign in, then exit

The install also inventories vision reachable through your other local harness CLIs (Codex, OpenCode, Pi) and asks, per harness, whether modlens may reuse it. Granted logins join the engine pool as equals, and every reused read is labeled with whose quota it spent.

On DeepSeek Harness the command line is not the only way in. Settings → Plugins → Plugin config carries a ModLens card (from dsh 0.1.7 on, open the sidebar's Plugins page and pick @liustack/modlens under Installed): switch the engine, tick which local CLIs auto mode may reuse, hit save and it takes effect.

[The ModLens vision-engine card in the dsh settings page, shown in Chinese: switch the engine, tick which local CLIs auto mode reuses]

Usage

Once installed, just chat. Paste an image or drop a path, ask anything, and the skill triggers on its own: the image goes to a vision engine and the answer comes back grounded in what it read. Paste once, and later questions about the same image do not need another paste.

Vision engines: six built-in providers, four reusable CLIs, one failover chain

ModLens does not depend on any single vision service. Ten sources of vision in total: six built-in providers, any one of which is enough, plus four local agent CLIs whose logins can be reused. The built-ins:

Without a pinned provider, every configured engine forms one failover chain: the fast API providers try first, the agent CLIs back them up, the first good result wins, and meta.attempts records every attempt so a fallback is never silent.

openai is a universal socket, not just OpenAI

Any endpoint speaking the OpenAI chat-completions protocol with image input plugs straight in — that covers most of the vision-model world:

modlens config set openai.baseUrl https://dashscope.aliyuncs.com/compatible-mode/v1   # qwen-vl
modlens config set openai.apiKey  <key>
modlens config set openai.model   qwen3-vl-plus

apiKey (and the matching env var) also accepts a comma-separated list. ModLens rotates to the next key after authentication, rate-limit, or quota failures. Network, 5xx, and parse failures skip remaining keys and keep provider failover.

The same three keys work for GLM's open platform, SiliconFlow, OpenRouter, a self-hosted vLLM/Ollama, or any gateway of your own. If your favorite vision model has an OpenAI-compatible API, ModLens can drive it.

Reusing what your machine already has

Two more sources of vision need zero new keys, each behind one explicit consent recorded in config:

• The harness you are talking in right now. Running inside Claude Code with a subscription signed in? claude-cli reads images through it out of the box. The install flow asks the same question for whichever harness you install into.
• Every other agent CLI on the machine. modlens doctor discovers them, you grant per harness, and they join the same failover chain with no priority over your own keys. Every reused read is labeled in meta.warnings with whose quota it spent, so nothing is ever silently billed:

Picking and routing

Two knobs: modlens config set provider <name> states a preference (the chain still backs it up), -p <name> pins exactly one with no fallback. Machines behind a proxy set HTTPS_PROXY or modlens config set proxy <url> and the API providers route through it. An internal endpoint can opt out with modlens config set openai.proxy "". Details: the CLI manual for defaults and flags, Configuration for every key, and Security for who fetches what on remote URLs.

See it work

Unedited runs, all driving a text-only DeepSeek-V4-Flash.

The newest one first: pasting a screenshot straight into DeepSeek Harness on the DeepSeek-V4-Flash (modlens vision) variant. The paste keeps its native thumbnail, the trajectory shows the image arriving "already transcribed by the modlens vision bridge", and the answer walks the UI element by element.

[Pasting an image straight into DeepSeek Harness, read through the modlens vision plugin]

A tweet screenshot in the Codex desktop app. It reads the author, the caption, the photo itself (down to what both people are wearing), the timestamp, and every engagement number: 5.4M views, 1.6K replies, 5.7K reposts, 116K likes.

[Text-only DeepSeek reading a tweet screenshot in full detail via ModLens]

Three images pasted at once. The model reads them one by one, spots that they belong to one visual family, and describes each illustration's content and style.

[Three images dropped together, read one by one]

The stress test: a scatter plot comparing 128 AI models. It reads both axes, the log scale, the per-provider color coding, the highlighted region, and every DeepSeek model called out with dashed markers. Dense charts are where vision bridges most often fail.

[The 128-model scatter plot read in full: axes, log scale, and highlighted region]

And the paste path, end to end, in a Claude Code terminal on DeepSeek. The pasted image arrives as a path rather than pixels, the skill triggers on its own, the guard confirms the model truly has no vision, and the slide's full content comes back: titles, layout, background, plus an honestly stated uncertainty about the truncated filename.

[The skill triggering on its own in a DeepSeek Claude Code session and reading a pasted slide]

Documentation

Contributing

ModLens does not accept pull requests. The project is maintained by a single author who reviews every line, which is a deliberate choice for reliability. Two effective ways to contribute:

• Open an issue. Bugs, suggestions, confusing errors, unclear docs. Issues are read and shape what gets built next.
• Fork it. Under MIT your copy is fully yours to modify and publish.

Shameless plug

ModSearch is ModLens's sibling project, the same craft applied to another missing sense: it gives models with no web access web search, X search, and single-page fetch. Free, no signup, no API key. A model that needs ModLens for its eyes usually needs ModSearch for the web:

npx -y @deepseek-ai/dsh plugin --profile web add @liustack/modsearch@latest

Follow the liustack WeChat official account: AI startup opportunities, indie-dev insights, and hands-on AI tooling, delivered as they happen. Scan the QR code in WeChat, or search for "liustack":

⭐ If it helps, star ModLens and ModSearch. Stars are how the next developer finds them.

Key ecosystem partners

The projects worth recommending in the DeepSeek Harness ecosystem.

• 🛒 dsh-market — The plugin market inside DeepSeek Harness. Browse 800+ community plugins with category filters and screenshot previews, one-click install and update, and live theme switching. Most need no restart. DeepSeek Harness 的可视化插件市场。设置页里直接逛社区全部 800+ 插件：分类筛选、截图预览、一键安装与更新、主题即点即换，装完多数免重启。
• 🖥️ DeepSeek Harness Desktop — A desktop front end for DeepSeek Harness. Start and manage the Harness service on your own machine without installing Node.js or running a command. A plugin market, remote control from a phone, and IM channels are on its roadmap. Site 为 DeepSeek Harness 生态打造的现代化桌面端。不用配置 Node.js，也不用敲命令，就能启动和管理本机的 Harness 服务。后续还会支持插件市场、移动端远程控制和 IM Channels。官网

Star History

Disclaimer

Provided as-is under the MIT License below. The author makes no warranty and gives no endorsement for any particular use, commercial use included. Your use of upstream engines (Antigravity CLI, the Gemini, OpenAI, and Anthropic APIs, and any OpenAI-compatible endpoint) is governed by their own terms and quotas, which you are responsible for.

License

MIT