niche-explorer
Skill0 starsChanged 7 months agoMITView on GitHub
Structured domain exploration that produces actionable market analysis and MVP recommendations
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- harness/CLAUDE.mdCLAUDE.md
- auto-security-scanSkill
- auto-test-runnerSkill
- compactSkill
- content-library-producerSkill
- content-publisherSkill
- fastapi-service-templateSkill
- frontend-designSkill
- human-review-gateSkill
- integration-testerSkill
- ios-agentSkill
- ios-designSkill
- mvp-spec-writerSkill
- overnight-dispatchSkill
- pwa-frontend-liteSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 40
Read the domain's CLAUDE.md and understand:
≈ “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
Niche Explorer Skill
Performs structured market exploration within a FORGE domain, analyzing existing projects, identifying gaps, and recommending MVPs with data-driven justification.
When to Use
- Starting work in a new domain
- Looking for the next MVP to build
- Validating market opportunities
- Filling gaps in domain coverage
Prerequisites
- Domain CLAUDE.md exists at
{domain}/CLAUDE.md - Access to web search for market research
- Domain context from living docs
Inputs
domain: babybit-es
focus: baby nutrition
constraints:
- COPPA compliant
- Parents as primary usersWorkflow
Step 1: Load Domain Context
Read the domain's CLAUDE.md and understand:
- Target market
- Existing projects and their status
- Domain-specific compliance requirements
- Technology stack preferences
Step 2: Analyze Existing Projects
For each project in the domain:
- Current status (complete, in-progress, docs-only)
- Coverage area (what problem it solves)
- Gaps (what it doesn't cover)
Step 3: Market Research
Conduct competitive analysis:
- Identify 3-5 top competitors
- Analyze their feature sets
- Find underserved segments
- Estimate market sizes
Step 4: Gap Analysis
Compare existing projects to market needs:
- What problems are unsolved?
- What segments are underserved?
- What adjacent opportunities exist?
- What compliance gaps exist?
Step 5: Generate Recommendations
Rank opportunities by:
- Market Size: Large (1M+ TAM), Medium (100K-1M), Small (<100K)
- Competition: Low (0-2 competitors), Medium (3-5), High (6+)
- Fit Score: 0.0-1.0 based on domain expertise, tech stack match, compliance readiness
Step 6: Output Structured Report
Generate exploration report following this template:
# Niche Exploration: {domain}
**Date:** {date}
**Focus:** {focus_area}
## Executive Summary
1-2 paragraph summary of findings and top recommendation.
## Market Analysis
### Target Segment
- Primary users: [description]
- Pain points: [bullet list]
- Current solutions: [what exists]
### Market Size
- TAM (Total Addressable Market): [estimate]
- SAM (Serviceable Addressable Market): [estimate]
- SOM (Serviceable Obtainable Market): [estimate]
## Existing Projects
| Project | Status | Coverage | Gaps |
|---------|--------|----------|------|
| project-1 | 80% complete | Feature A, B | Missing C, D |
| project-2 | Docs only | N/A | Not started |
## Competitive Landscape
| Competitor | Strengths | Weaknesses | Market Share |
|------------|-----------|------------|--------------|
| Competitor A | Feature X, Y | Poor UX | 35% |
| Competitor B | Great design | Missing Z | 20% |
## Gap Analysis
### Opportunity 1: {Name}
- **Problem**: [What pain point does this solve?]
- **Solution**: [High-level approach]
- **Market Size**: Large/Medium/Small
- **Competition**: Low/Medium/High
- **Fit Score**: 0.85
- **Why Now**: [Market timing, trends]
### Opportunity 2: {Name}
[Same format]
## Recommended MVP
**{MVP Name}** - {One-line description}
### Problem Statement
[Detailed description of the problem this solves]
### Target Users
- Primary: [user persona]
- Secondary: [user persona]
### Solution Overview
[How this MVP solves the problem]
### Differentiator
[What makes this unique vs competitors]
### Success Metrics
- Metric 1: [e.g., 100 active users in 30 days]
- Metric 2: [e.g., 70% user retention]
- Metric 3: [e.g., 4.5+ app store rating]
### Tech Stack Fit
- Backend: [FastAPI / Go / Node.js]
- Frontend: [React PWA / Lit / HTMX]
- Auth: [JWT / OAuth2]
- Database: [PostgreSQL / MongoDB]
- Compliance: [COPPA / HIPAA / GDPR requirements]
### Estimated Effort
- Complexity: Low/Medium/High
- Timeline: [X weeks]
- P0 Features: [count]
## Next Steps
1. Run `/mvp-spec-writer` to generate features.json
2. Create features.json with 5-7 P0 features
3. Run `forge loop run` to start building
4. Track in Command Center
---
**Output Path**: `{domain}/explorations/{mvp-name}-exploration.md`Quality Criteria
A good exploration includes:
- At least 3 competitors analyzed
- Market size estimates with reasoning
- Clear differentiation from competitors
- Specific target user personas
- Measurable success metrics
- Tech stack alignment with domain standards
- Compliance requirements identified
- Fit score justification
Example Invocation
/niche-explorer
Domain: babybit-es
Focus: baby nutrition tracking
Constraints:
- COPPA compliant (parents are account owners)
- Integrates with existing meal-planner project
- Target Spanish-speaking familiesRelated Skills
/mvp-spec-writer- Convert exploration to features.json/living-docs- Update domain strategy with findingsforge loop run- Start building the MVP
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.

