[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-81382":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":16,"stars30d":17,"stars90d":15,"forks30d":15,"starsTrendScore":18,"compositeScore":19,"rankGlobal":10,"rankLanguage":10,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":21,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":15,"starSnapshotCount":15,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},81382,"mind-expander","mbbill\u002Fmind-expander","mbbill","An infinite-canvas workspace for coding with AI agents.","",null,"TypeScript",92,4,1,0,5,56,15,2.1,"Apache License 2.0",false,"main",[],"2026-06-12 02:04:14","\u003Cdiv align=\"center\">\n  \u003Ch1>mind-expander\u003C\u002Fh1>\n  \u003Cp>\n    \u003Cstrong>An infinite-canvas workspace for coding with AI agents.\u003C\u002Fstrong>\n  \u003C\u002Fp>\n  \u003Cp>\n    mind-expander turns a codebase into an infinite-canvas workspace for\n    developers coding with AI agents. Instead of reading another long\n    explanation, you and the agent can work on the same source-backed\n    graph: modules, types, calls, dependencies, ownership-like\n    relationships, and the parts of the system changed by a refactor or\n    PR. It supports Rust and TypeScript today, with more language\n    frontends planned. The goal is a better workspace for understanding\n    architecture, planning changes, reviewing PRs, and steering code\n    that is increasingly written with AI.\n  \u003C\u002Fp>\n  \u003Cp>\n    \u003Ca href=\"#why-it-exists\">Why It Exists\u003C\u002Fa>\n    ·\n    \u003Ca href=\"#how-to-use-it\">How To Use\u003C\u002Fa>\n    ·\n    \u003Ca href=\"docs\u002Fdemo-videos.md\">Demos\u003C\u002Fa>\n  \u003C\u002Fp>\n  \u003Cp>\n    \u003Ca href=\"https:\u002F\u002Fwww.npmjs.com\u002Fpackage\u002Fmind-expander\">\u003Cimg alt=\"npm version\" src=\"https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fv\u002Fmind-expander?style=flat-square&color=2563eb\">\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmbbill\u002Fmind-expander\u002Factions\u002Fworkflows\u002Frelease.yml\">\u003Cimg alt=\"release\" src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Factions\u002Fworkflow\u002Fstatus\u002Fmbbill\u002Fmind-expander\u002Frelease.yml?style=flat-square&label=release\">\u003C\u002Fa>\n    \u003Ca href=\"LICENSE\">\u003Cimg alt=\"license\" src=\"https:\u002F\u002Fimg.shields.io\u002Fnpm\u002Fl\u002Fmind-expander?style=flat-square&color=7c3aed\">\u003C\u002Fa>\n  \u003C\u002Fp>\n\u003C\u002Fdiv>\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0bb41ece-a493-4596-9c34-22d9c4e6dbea\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"docs\u002Fdemo-videos.md\">\u003Cstrong>Browse all demo videos\u003C\u002Fstrong>\u003C\u002Fa>\n  —\n  code maps · caller\u002Fcallee chains · diff review · planning tours\n\u003C\u002Fp>\n\n## Why It Exists\n\nCode review and design work are rarely about one file. The hard part is\nusually the shape of the system: who owns what, which API boundary is\nmoving, which function calls which, and whether a change crosses the\nwrong module boundary.\n\nThat is especially true when an AI agent is involved. Agents are good at\nproducing code and long explanations, but long markdown is a poor medium\nfor checking architecture. You need to see the same structure the agent\nis talking about, inspect the source behind each claim, and decide\nwhether the plan is actually right.\n\nmind-expander gives you that infinite-canvas surface. It extracts\nstructural facts from Rust and TypeScript, renders them as an\ninteractive browser diagram, and lets an agent turn a review, plan, or\nwalkthrough into a source-backed tour.\n\n## How To Use It\n\nmind-expander is designed to be driven by an AI coding agent.\n\n1. **Install the skill in your agent.** Copy this into your AI coding\n   agent:\n\n   ```text\n   Install the mind-expander skill by downloading the raw file at:\n   https:\u002F\u002Fraw.githubusercontent.com\u002Fmbbill\u002Fmind-expander\u002Frefs\u002Fheads\u002Fmain\u002Fskill\u002Fmind-expander.md\n\n   Install the exact downloaded contents as the skill file. Do not\n   summarize, rewrite, or paraphrase the file.\n   ```\n\n   If you prefer a local installer, run:\n\n   ```sh\n   npx mind-expander install-skill\n   ```\n\n\n2. **Ask naturally.** Examples:\n\n   > Use mind-expander to walk me through this codebase.\n\n   > Show me the architecture of this feature on the diagram.\n\n   > Review my current changes visually and explain the important relationships.\n\n   > Turn your refactor plan into an interactive tour.\n\n   > Walk me through the last few commits and show me where they matter.\n\n3. **The agent runs `npx mind-expander`** in the background — no\n   install step on your part. It opens a local diagram in your browser,\n   anchored to the source, narrated step by step.\n\n4. **You stay in control.** Click anything in the diagram to navigate,\n   open source in the side panel, or ask a follow-up question to spawn\n   a new walkthrough without losing context.\n\n## Typical Use Cases\n\n- **AI-guided PR review.** Open a diff view and ask your agent to walk\n  through the changed entities in structural context instead of reading\n  isolated hunks.\n- **Refactor planning.** Turn an agent's plan into an interactive tour\n  where every step points to the source it depends on.\n- **Codebase orientation.** Give yourself or a teammate a guided map of\n  the important modules, types, functions, and relationships.\n- **Architecture discussions.** Use the same visual context while\n  deciding where behavior belongs or how a boundary should move.\n- **Language-aware details.** Use Rust ownership\u002Flifetime signals and\n  TypeScript class\u002Finterface\u002Fmodule relationships where available, with\n  more language frontends planned.\n\n## CLI\n\nDriving mind-expander by hand instead of through an agent. Requires\nNode.js 18+.\n\n```sh\n# Open any workspace.\nnpx mind-expander view \u002Fpath\u002Fto\u002Fworkspace\n\n# Open with a diff overlay.\nnpx mind-expander view \u002Fpath\u002Fto\u002Fworkspace --at main..\n\n# List running instances + their pids.\nnpx mind-expander list\n```\n\n### Build from source\n\nFor unsupported platforms (musl Linux, BSD) or developing on the tool\nitself:\n\n```sh\ncargo install --git https:\u002F\u002Fgithub.com\u002Fmbbill\u002Fmind-expander\n```\n","mind-expander 是一个无限画布工作空间，用于与AI代理协同编程。该项目支持将代码库转换为一个无限画布的工作区，使开发者能够与AI代理在同一源代码支持的图表上进行协作，该图表包括模块、类型、调用、依赖关系等结构信息。目前支持Rust和TypeScript语言，并计划增加更多语言的支持。通过这种可视化的方式，可以更好地理解系统架构、规划变更、审查PR以及指导由AI生成的代码。适用于需要深入理解和管理复杂代码库的场景，尤其是在使用AI辅助开发时，帮助提高代码质量和开发效率。",2,"2026-06-11 04:04:50","CREATED_QUERY"]