spark-prompt
Skill0 starsChanged 5 months agoMITView on GitHub
Generate a GitHub Spark app prompt from an app idea. Use when: building a Spark app, creating a no-code app prompt, describing app requirements for GitHub Spark, drafting a single-prompt app description, or iterating on an existing Spark prompt. Produces a vibrant, feature-rich prompt ready to paste into Spark.
In this repository7 other files it gives its agents.
- antigravity-migrationSkill
- audio-normalizeSkill
- copilot-migrationSkill
- meeting-syncSkill
- meeting-to-documentationSkill
- meeting-transcribeSkill
- translate-docsSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Read the user's argument or the conversation.
- Distinctive: line 37
Do not ask about anything already evident from the idea.
- Distinctive: line 43
Use the template below as a structural guide.
- Distinctive: line 47
Use evocative, concrete language: *"a pulsing live feed"*, *"drag-and-drop kanban lanes"*, *"a glanceable at-a-glance dashboard"*
- Distinctive: line 62
Before presenting, run through this checklist:
- Distinctive: line 78
Ask if any feature, tone, or scope adjustment is needed
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
GitHub Spark Prompt Generator
Generate a compelling, feature-rich prompt for GitHub Spark — a no-code app builder that creates entire apps from a single prompt.
What This Produces
A single, self-contained prompt that:
- Opens with a vivid, specific description of the app's purpose
- Enumerates features as user-facing capabilities and visible elements
- Conveys the desired look, feel, and UX personality
- Ends with an extensibility hook for easy future iteration
- Does not include: file/folder structure, coding standards, implementation instructions, framework choices, or README requests
Procedure
Step 1 — Gather the App Idea
Read the user's argument or the conversation. If the app idea is clear and specific, proceed to Step 2. If it's vague or missing key details, ask only the questions that are genuinely unanswered:
| Gap | Question to ask |
| Purpose unclear | What is the single most important thing this app does? |
| Audience unclear | Who uses it, and in what context (personal, team, public)? |
| Data undefined | What does the app store, track, or display? |
| Interactions missing | What are the key actions a user takes? |
| Vibe undefined | What's the visual personality? (e.g., minimalist, playful, data-dense, bold) |
| Scope unclear | Is this focused and single-purpose, or a small multi-feature tool? |
Do not ask about anything already evident from the idea.
Step 2 — Draft the Spark Prompt
Use the template below as a structural guide. Write in flowing, specific prose — avoid generic phrases like "user-friendly" or "intuitive".
DO:
- Open with a punchy one-sentence elevator pitch naming the app and its purpose
- Use evocative, concrete language: "a pulsing live feed", "drag-and-drop kanban lanes", "a glanceable at-a-glance dashboard"
- Describe everything from the user's perspective: "users can…", "the app surfaces…", "tapping X reveals…"
- Include a visual/UX direction: color mood, layout style, motion character
- Close with an extensibility statement tailored to the app's natural growth axes
DON'T:
- Mention filenames, directories, or project layout
- Specify languages, frameworks, libraries, or patterns
- Give implementation instructions ("use a REST API", "persist in localStorage")
- Request a README or any documentation
Step 3 — Validate
Before presenting, run through this checklist:
- Is the core purpose immediately clear in the first sentence?
- Are all features described as user actions or visible elements, not technical choices?
- Is the language vivid and specific, not generic?
- Is there a clear UX/visual direction?
- Does the prompt end with an extensibility hook?
Revise any section that fails a check.
Step 4 — Present and Offer to Iterate
- Deliver the final prompt in a fenced code block for easy copying
- Note 1–2 assumptions made (scope, tone, key feature prioritization)
- Ask if any feature, tone, or scope adjustment is needed
- Offer to write a follow-up iteration prompt once the first Spark version is built
Prompt Template
[App Name] is a [vivid one-sentence description of what the app does and who it's for].
**Core Experience**
[2–3 sentences on the primary workflow and what makes it feel great to use. Be specific about layout, interactions, and visual energy.]
**Features**
- [Feature 1: user capability or visible element]
- [Feature 2]
- [Feature 3]
- [Feature 4]
- [Feature 5+]
**Look & Feel**
[1–2 sentences on color palette mood, typography style, motion/transition character, and overall visual personality.]
**Extensibility**
The app should be designed so that [specific growth axis — e.g., new categories, additional views, or extra data fields] can be added with minimal friction as the tool evolves.Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

