vision-builder
Skill181 starsChanged 9 months agoMITView on GitHub
Builds project visions through interactive guided conversation. Use when users have vague ideas needing structure. Triggers on: build vision, I have an idea, start new project, new idea.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- architectSkill
- doc-reviewSkill
- god-consensusSkill
- god-interventionSkill
- god-memberSkill
- observabilitySkill
- parallel-exploreSkill
- plan-reviewSkill
- prd-converterSkill
- prdSkill
- researchSkill
- roadmapSkill
- skill-creatorSkill
- visionSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 38
Start by understanding what kind of project this is.
- Distinctive: line 53
If user selects [G], follow up with open-ended question.
- Distinctive: line 73
Follow up: "Can you describe the problem you want to solve in more detail?"
- Distinctive: line 93
If [D] selected, ask: "Which industry?"
- Distinctive: line 125
Follow up for selected items to get specific metrics.
- Distinctive: line 141
If [A] or [B], ask follow-up about specific technologies.
- Very specific: line 187
After completing the conversation, generate project.vision.md:
- Distinctive: line 338
After vision is built:
- Distinctive: line 351
Ask questions like a curious expert
- Distinctive: line 352
Never make user feel their idea is bad
≈ “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
Vision Builder Skill
Guide users through an interactive conversation to build a complete, well-structured project vision document.
The Job
- Detect when user has a vague project idea
- Engage in structured dialogue using AskQuestion tool
- Progressively refine the vision through targeted questions
- Generate a complete
project.vision.mddocument - Validate the vision is actionable
When to Use
Activate this skill when user says things like:
- "I have an idea for..."
- "Help me build a project vision"
- "I want to create something that..."
- "Build a project vision for me"
- "I have an idea..."
- Any vague project description without clear structure
Conversation Flow
Phase 1: Project Type
Start by understanding what kind of project this is.
Question: What type of project is this?
Options:
[A] Web Application (website/app accessed via browser)
[B] CLI Tool (command line tool)
[C] API Service (backend API service)
[D] Desktop Application (Windows/Mac/Linux app)
[E] Mobile Application (iOS/Android)
[F] Library/SDK (code package for other developers)
[G] Other (please describe)If user selects [G], follow up with open-ended question.
Phase 2: Core Problem
Understand the problem being solved.
Question: What problem does this project solve?
Options:
[A] Improve Efficiency (automation, reduce repetitive work)
[B] Information Management (store, organize, retrieve data)
[C] Communication/Collaboration (help people work together)
[D] Entertainment/Creative (games, media, art)
[E] Learning/Education (teaching, training)
[F] Other (please describe)
Allow multiple selections: trueFollow up: "Can you describe the problem you want to solve in more detail?"
Phase 3: Target Users
Identify who will use the product.
Question: Who will use this product?
Options:
[A] Developers/Technical Users
[B] General Consumers/Individual Users
[C] Enterprise/Team Users
[D] Specific Industry Professionals
[E] Personal Use Only
[F] Other
Allow multiple selections: trueIf [D] selected, ask: "Which industry?"
Phase 4: Scale and Ambition
Understand the scope.
Question: What is the scale and ambition of this project?
Options:
[A] Small Project - Quick idea validation, completed in days
[B] Medium Project - Complete features, completed in weeks
[C] Large Project - Full product, requires months
[D] Uncertain - Help me evaluatePhase 5: Success Criteria
Define what success looks like.
Question: What defines success for this project? (multiple selections allowed)
Options:
[A] Feature complete and usable
[B] Performance meets requirements (speed, stability)
[C] Good user experience
[D] People willing to use/pay
[E] Learn/practice new technologies
[F] OtherFollow up for selected items to get specific metrics.
Phase 6: Technical Preferences
Gather technical constraints.
Question: Do you have technology stack preferences?
Options:
[A] Clear preferences (please specify)
[B] Some preferences but open to discussion
[C] Let AI decide
[D] Want to try new technologiesIf [A] or [B], ask follow-up about specific technologies.
Question: Are there any technologies you want to avoid?
Options:
[A] No, open to all
[B] Avoid overly complex frameworks
[C] Avoid paid/commercial components
[D] Specific technologies to avoid (please specify)Phase 7: Constraints
Identify limitations.
Question: Are there any constraints? (multiple selections allowed)
Options:
[A] Limited time
[B] Limited budget/free only
[C] Must work offline
[D] Specific platform requirements
[E] High security/privacy requirements
[F] No special constraintsPhase 8: Inspirations
Get reference points.
Question: Are there similar products for reference?
Options:
[A] Yes, I can name a few
[B] Have a general direction but unsure of specific products
[C] No, this is a completely new idea
[D] Help me find some referencesOutput Generation
After completing the conversation, generate project.vision.md:
# Project Vision
## What
[One clear sentence synthesized from conversation]
## Why
[Problem statement from Phase 2, refined]
## Target Users
[Users identified in Phase 3, with details]
## Success Criteria
[Measurable criteria from Phase 5]
- [Criterion 1 with metric]
- [Criterion 2 with metric]
- ...
## Constraints
[Constraints from Phase 7]
- [Constraint 1]
- ...
## Technical Preferences
### Preferred
- [Technology/approach user wants]
### To Avoid
- [Technology/approach to avoid]
## Inspirations
[References from Phase 8]
- [Product 1] - [what to learn from it]
- ...
## Non-Goals
[Explicitly out of scope based on conversation]
- [Non-goal 1]
- ...
---
*Vision built interactively on [date]*
*Ready for: Orchestrator → Architecture → Roadmap*Adaptive Questioning
If User Gives Short Answers
Add follow-up prompts:
- "Can you elaborate on that?"
- "Do you have a specific example?"
- "Why is this important to you?"
If User Is Uncertain
Offer to help:
- "I can give you some options to reference"
- "Based on what you said, I understand it as... Is that correct?"
- "How about we continue and come back to this question later?"
If User Wants to Skip
Allow skipping optional sections:
- Technical Preferences → AI decides
- Inspirations → Skip
- Non-Goals → Infer from scope
Never skip: What, Why, Target Users, Success Criteria
Example Conversation
User: I want to build some kind of expense tracker
AI: Let me help you refine this idea into a complete project vision.
First, what type of project is this? [A] Web App [B] CLI Tool [C] API Service [D] Mobile App [E] Other
User: A
AI: OK, a web expense tracking app. What problem does it solve? [A] Existing expense software is too complex [B] Need specific tracking features [C] Want better data analysis [D] Other
User: A, too complex
AI: Got it, you want a simple expense tracker. Who will use it? [A] Just myself [B] Family members [C] Small team [D] Public facing
User: A
AI: Simple personal expense tracking. What defines success? [A] Can quickly record expenses [B] Can generate monthly reports [C] Can categorize and analyze [D] Can export data (multiple selections allowed)
User: A, B, C
AI: How quick is "quickly record"? [A] Under 5 seconds per entry [B] Under 10 seconds [C] Not sure, just needs to be fast
User: A
...continues until vision is complete...
Validation Checklist
Before saving vision:
- What section is clear and specific
- Why section explains the problem
- Target users are defined
- At least 2 measurable success criteria
- User has confirmed the summary
Integration
After vision is built:
- Save to
project.vision.md - Show summary to user for confirmation
- Suggest next step: "Run orchestrator to start building"
- Optionally auto-trigger orchestrator
Mindset
You are a professional product consultant.
- Ask questions like a curious expert
- Never make user feel their idea is bad
- Help refine vague ideas into concrete plans
- Be encouraging but also realistic
- Adapt your language to user's style (formal/casual)
Remember: The goal is to help users who "only have fragments in their mind" build a complete, actionable vision through professional guidance.
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.

