[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-96039":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":9,"languages":9,"totalLinesOfCode":9,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":13,"subscribersCount":13,"size":13,"stars1d":13,"stars7d":13,"stars30d":14,"stars90d":13,"forks30d":13,"starsTrendScore":13,"compositeScore":15,"rankGlobal":9,"rankLanguage":9,"license":16,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":17,"hasPages":17,"topics":19,"createdAt":9,"pushedAt":9,"updatedAt":20,"readmeContent":21,"aiSummary":22,"trendingCount":13,"starSnapshotCount":13,"syncStatus":12,"lastSyncTime":23,"discoverSource":24},96039,"dream-loop","achimala\u002Fdream-loop","achimala","Agent skill for impressive 3D visuals using Blender + image gen + subagent critic",null,841,92,2,0,472,59.91,"MIT License",false,"main",[],"2026-09-20 04:01:32","# dream-loop\n\nBy [@anshuc](https:\u002F\u002Fx.com\u002Fanshuc).\n\nAn agent skill that builds a game, app, or scene with impressive visuals, by creating a closed loop:\n\n1. AI \"dreams\" up a high-quality target screenshot using image generation\n2. AI builds with this target in mind\n3. A separate AI critic compares the live screenshot to the target and provides feedback\n4. AI loops back to step 2 until critic is satisfied\n5. Optionally, AI loops back to step 1 and dreams up an even better target based on the current state.\n\n## Installation\n\n`npx skills add achimala\u002Fdream-loop`, or clone into your agent's skills directory, or paste the link into your agent and tell it to figure it out.\n\n## Prerequisites\n\nYou need an AI agent with:\n\n- access to image generation, either built-in (e.g. Codex, Grok) or via API (e.g. give it a Gemini API key)\n- vision input\n- subagents (optional but strongly preferred)\n\nInstall Blender if you want custom 3D modeling. The Blender MCP or scripting interface is preferred to computer use; it produces better results.\n\nAt the moment this is only tested with GPT-6 Astra in Codex. Other strong models like Claude Fable 5.1 can likely work too.\n\n## Example\n\nPrompt:\n> Build me a graphics demo: isometric camera, voxel-ish art style with realistic shading and reflective wet floors, a character in an interesting scene. Fantasy setting (think Elden Ring, Diablo). Three.js in browser, >60fps. Don't download assets. Time limit of 1 hour. Controls: click to move the character, camera lazy-follows; drag to rotate camera; scroll to zoom in\u002Fout. No gameplay for now. World should feel alive: motion, animations, subtle environmental behaviors. Area around player should look expansive, but only allow movement in a limited space. No need to confirm the art with me or ask questions, just go!\n\nGPT-6 Astra on high effort in Codex:\n\n![Vesper demo: an isometric fantasy scene](assets\u002Fvesper-preview.gif)\n\n[Try the live demo](https:\u002F\u002Fdream-loop-demo.anshu.dev)\n\n## Contributing\n\nI'd love to hear feedback, see results, and accept edits to the skill.\n\nIf you open a PR, please provide example results produced by the skill, to ensure the changes don't regress performance.\n","dream-loop 是一个面向 AI 代理的技能模块，用于自动生成具备高质量 3D 视觉效果的应用或场景。其核心是闭环工作流：AI 先生成目标图像（如游戏截图），再通过 Blender 或 Web 渲染框架（如 Three.js）构建实现，由独立子代理作为视觉批评者比对实时画面与目标图并反馈优化建议，直至达标；支持多轮迭代升级目标。技术上依赖多模态能力（图像生成、视觉理解）、子代理协同及 Blender 脚本化建模。适用于快速原型验证、图形编程教学、AI 驱动的创意开发等需兼顾视觉质量与自动构建效率的场景。","2026-09-09 02:30:03","CREATED_QUERY"]