1. Conduid
  2. Files
  3. Pi Vision Struct
MCP server · Files

Pi Vision Struct

⛔ 已停止维护(EOL 2026-09-13,仓库只读)· Deprecated & archived:给纯文本模型的数理化视觉通道 + GUI 控制;已被 DeepSeek-V4.1 原生多模态取代。Numeric-grade vision channel + GUI control for text-only models — superseded by native multimodal DeepSeek-V4.1.

34Low

Scored 27 days ago · breakdown

About Pi Vision Struct

Pi Vision Struct is an MCP server published by Moritz230127 in the Files category: ⛔ 已停止维护(EOL 2026-09-13,仓库只读)· Deprecated & archived:给纯文本模型的数理化视觉通道 + GUI 控制;已被 DeepSeek-V4.1 原生多模态取代。Numeric-grade vision channel + GUI control for text-only models — superseded by native multimodal DeepSeek-V4.1.

The repository has 5 stars and 0 forks, with the last commit a month ago. Recent commit activity is the strongest signal that reported issues will get attention.

Install

From source
git clone https://github.com/Moritz230127/pi-vision-struct

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Pi Vision Struct

Powered by Claude · Grounded in docs

I know everything about Pi Vision Struct. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Releases

v4.0.3v4.0.3 — 项目终止维护 / End of Life · 13 Sep 2026⛔ 项目终止维护 / End of Life 本仓库已于 2026-09-13 正式封库,不再维护与更新。** 中文 决定 `pi-version-and-gui` 项目自即日起停止一切维护与更新,仓库归档为**只读**, 不再接受 issue 与 PR。这是本项目的最后一次公告。 原因 本项目的核心前提是「纯文本模型需要外部视觉通道」——把图片转化为坐标 / hex / 矩阵 / 统计等精确数理数据,供文本模型做数值推理与 GUI 控制。 DeepSeek-V4.1…
v4.0.0V4.0.0 — 全面 Rust + GUI 控制 · 8 Sep 2026V4.0.0 发布:全面 Rust 化 + GUI 控制 37 个传感器/模块全 Rust(零 Python 回退) GUI 控制:a11y_action 直达 + 五级降级链 + 闭环验证 241 测试全绿 + clippy 0 + 差分门禁 12 个 性能:scene_stats 17x / pix 4.1x / ocr 1.8x GUI 延迟:capture 40ms / analyze 226ms / click 216ms
v3.0.0v3.0.0: D-S 融合引擎 + 轻量 DL 传感器 + 调度管理器 + schema v3 · 1 Sep 2026v3.0.0 — V3 大版本(2026-09-01) ① 融合引擎(核心创新) `vs_fusion.py`:D-S 证据理论融合(mass 映射 + 正交和组合 + belief/plausibility/uncertainty)+ 匈牙利全局匹配 替代旧 `vs_crosscheck.py` 硬阈值:边界场景误报率 70% → 0%(实证) ② 6 个新传感器 `vs_saliency.py`:U²-Net(ONNX GPU)显著性 → top-N…
v2.2.1v2.2.1: 修复 analyze 传自然语言指令崩溃 · 29 Aug 2026问题 用户调用 `analyze` 时把**自然语言指令**(如"分析这张截图中的 fcitx5 候选框外观…")直接传给 `--task`,而 `vs_analyze.py` 把 task 当作配置文件名去打开 `tasks/<task>.json` → `[Errno 36] File name too long` 崩溃,经 bwrap 沙箱包装后表现为 `python failed`。 修复 `vs_analyze.py`:task…
v2.2.0v2.2.0: 真几何深度 + VLM 语义兜底 + 依赖健壮性 · 29 Aug 2026本次更新 ① 真几何深度(depth_geom)—— 不再是亮度梯度 新增 `vs_depth_geom.py` + `depth_geom_wrapper.sh`:在 Blender 内将所有 MESH 三角面栅格化到相机空间深度缓冲,输出**真实几何距离(米)** 向量化投影 + numpy 扫描线光栅化:480p ≈ 42s / 1080p ≈ 57s 涡扇实测:near=0.75m / far=2.05m / median≈1.9m(与引擎真实尺寸吻合,非亮度近似)…
Questions

About Pi Vision Struct

How do I install Pi Vision Struct?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Pi Vision Struct safe to use with an AI agent?

Its trust score is 34 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Pi Vision Struct still maintained?

Yes — the latest release is v4.0.3 (13 Sep 2026), and the last commit was a month ago. The repository has 5 stars.