Skip to content
Back to search

dream-loop

SkillWorks with Claude CodeGitHub

Build a game or app from a description so that a live screenshot matches a generated rendering. Use when the user says "dream loop" or asks for something built to a very high level of graphical fidelity.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/achimala-dream-loop/badge.svg)](/listings/achimala-dream-loop)

HTML

<a href="/listings/achimala-dream-loop"><img src="/api/listings/achimala-dream-loop/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

`npx skills add achimala/dream-loop`, or clone into your agent's skills directory, or paste the link into your agent and tell it to figure it out.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 49 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: dream-loop
  3. description: Build a game or app from a description so that a live screenshot matches a generated rendering. Use when the user says "dream loop" or asks for something built to a very high level of graphical fidelity.
  4. license: MIT
  5. ---
  6. # dream-loop
  7. This is a process for you to autonomously build extremely impressive visuals, especially for 3D scenes (e.g. in a game or app).
  8. Your first step is to determine which workflow to use:
  9. - If the user told you to use the Plus or Pro workflow explicitly, that's your answer
  10. - If not, check if the user is on a low or high tier coding subscription. "Low tier" means ChatGPT Plus or equivalent. "High tier" means ChatGPT Pro. For low-tier subscriptions, use the Plus workflow. For high-tier, use the Pro workflow.
  11. Plus workflow: read [references/plus-mode/workflow.md](references/plus-mode/workflow.md)
  12. Pro workflow: read [references/pro-mode/workflow.md](references/pro-mode/workflow.md)
  13. Do not read both documents. They are not inter-compatible.
  14. # Guidance applicable to both workflows
  15. ## General
  16. Create and use a `.dream-loop` folder for working context/files, and gitignore it.
  17. ## The target image
  18. The key piece of Dream Loop is to first create the "dream version" of the user request using image generation, then iterate to build it.
  19. If the user supplies this, use that directly.
  20. If not, you need to generate it.
  21. If you don't have an image generation tool, stop and ask the user to either provide the target image, or connect you to an image generation API.
  22. Before generating the image, determine if there is an existing product or if you're starting fresh. If fresh, you can directly generate a new image. If there's an existing product, you should capture a screenshot of the current version of it, then use that as the baseline input to the image model and generate a refined version of it based on the user direction, so that it is an improvement over the original and not a divergence.
  23. When generating images, avoid using words like "concept art" in the prompt. This is not an artist's interpretation. It is meant to be an exact, realistic target screenshot. You will try to match it down to the pixel. For example, if the user is asking you to make a game, you should prompt the image model for a real in-engine screenshot of the target. Not a cinematic shot, photo, painting, artist concept, etc.
  24. Store the image in `.dream-loop/target.png`.
  25. ## Time budget
  26. If the user gives a time budget, record the time at start of the loop (after locking target.png), and check the clock between rounds.
  27. Don't degrade visual fidelity to hit the time budget. Don't take shortcuts. It's better to hit the time limit with meaningful, beautiful progress than with something roughly complete but ugly.
  28. If the user doesn't give a time budget, run until you hit an exit criterion, but warn that this may consume a lot of tokens.
README.md · README · 44 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # dream-loop
  2. By [@anshuc](https://x.com/anshuc).
  3. An agent skill that builds a game, app, or scene with impressive visuals, by creating a closed loop:
  4. 1. AI "dreams" up a high-quality target screenshot using image generation
  5. 2. AI builds with this target in mind
  6. 3. A separate AI critic compares the live screenshot to the target and provides feedback
  7. 4. AI loops back to step 2 until critic is satisfied
  8. 5. Optionally, AI loops back to step 1 and dreams up an even better target based on the current state.
  9. ## Installation
  10. `npx skills add achimala/dream-loop`, or clone into your agent's skills directory, or paste the link into your agent and tell it to figure it out.
  11. ## Prerequisites
  12. You need an AI agent with:
  13. - access to image generation, either built-in (e.g. Codex, Grok) or via API (e.g. give it a Gemini API key)
  14. - vision input
  15. - subagents (optional but strongly preferred)
  16. Install Blender if you want custom 3D modeling. The Blender MCP or scripting interface is preferred to computer use; it produces better results.
  17. At the moment this is only tested with GPT-6 Astra in Codex. Other strong models like Claude Fable 5.1 can likely work too.
  18. ## Example
  19. Prompt:
  20. > 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/out. 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!
  21. GPT-6 Astra on high effort in Codex:
  22. ![Vesper demo: an isometric fantasy scene](assets/vesper-preview.gif)
  23. [Try the live demo](https://dream-loop-demo.anshu.dev)
  24. ## Contributing
  25. I'd love to hear feedback, see results, and accept edits to the skill.
  26. If you open a PR, please provide example results produced by the skill, to ensure the changes don't regress performance.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.