# liustack/modlens repository brief

> This is an EasyGlobe-authored bilingual repository brief, not a redistributed third-party Skill and not an installable copy of the repository. Review the upstream repository before use.

## English

The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding agent. Paste an image, get structured JSON evidence (OCR, layout, semantics). | 全网最强 DeepSeek Harness 外挂视觉插件，为 DeepSeek、GLM 等纯文本模型外挂视觉能力，粘贴图片即得结构化 JSON 证据（OCR、版面、语义）

EasyGlobe verified this public, non-fork, non-archived repository at commit `aaeb5664290f8ffa701261bce3284edbaa21a5d8`. The sampled tree contains 1 `SKILL.md` file and at least one sampled file has valid `name` and `description` frontmatter. Its primary directory category is Creative & Design. Static sampling did not find the high-risk credential-exfiltration or destructive-root-delete patterns checked by the research pipeline; this is not a full security audit.

## 中文

liustack/modlens 是一个面向创意与设计工作流的 GitHub Skill 仓库。EasyGlobe 在提交 `aaeb5664290f8ffa701261bce3284edbaa21a5d8` 对公开、非 fork、未归档的默认分支进行了抽样核验：仓库树包含 1 个 `SKILL.md` 文件，且至少一个抽样文件具有有效的 `name` 与 `description` frontmatter。静态抽样未命中本流水线检查的凭据外传或根目录破坏性命令模式；这不等同于完整安全审计，使用前仍应检查上游仓库。

## Repository metadata

| Field | Value |
| --- | --- |
| Repository | [liustack/modlens](https://github.com/liustack/modlens) |
| Hot rank / score | 68 / 46.355 |
| Stars at verification | 3427 |
| Default branch | `main` |
| Sampled commit | `aaeb5664290f8ffa701261bce3284edbaa21a5d8` |
| Last push | 2026-08-20T20:05:55Z |
| Verified | 2026-08-21T03:25:24.715Z |
| License | MIT |
| SKILL.md paths in sampled tree | 1 |

## Representative valid Skill evidence

| Path | Frontmatter name | Frontmatter description | SHA-256 |
| --- | --- | --- | --- |
| `skills/modlens/SKILL.md` | modlens | Plug-in vision for text-only models. Hard rule: when a file path or URL with an image extension (.png, .jpg, .jpeg, .webp, .gif, .heic, .heif) appears anywhere in the conversation (typed by the user, injected as a `[Image: source: <path>]` line, or inside a tag) and you cannot see that image's content, run this skill on it before any other approach: no self-built OCR, no PIL, no tesseract. Also triggers on pasted-image placeholders such as `[Image #1]` and `[Unsupported Image]`. If you can actually see the image, do not use this skill. When unsure, run `modlens guard` before the first read of a session: a deny verdict means the active model has native vision and must read the image itself. Runs the modlens CLI to convert the image into structured JSON evidence: every word transcribed, layout regions, semantics, visual clues. Also use when the user asks how to install, configure, or switch modlens providers (Gemini API key, OpenAI-compatible endpoints, Claude API or Claude Code CLI). | `bf38f78cfc57b1546a8191b153e8bb47ead8600dcfae672f1f72d3a27d9ebec7` |

Source: https://github.com/liustack/modlens