fdl-mcp-server
# fdl-mcp-server
> Model Context Protocol server for **fivedaylaunch.com** — lets Claude (and any MCP-compatible AI client) audit, compare, and redesign any small business website. Free, public, no API key required.
[](https://opensource.org/licenses/MIT)
[](https://modelcontextprotocol.io)
## What this does
Five tools that any MCP client can invoke:
| Tool | What it does |
|---|---|
| `audit_website` | Deterministic 5-pillar audit (Performance, SEO, Mobile, Security, AEO) → score 0-100 + specific findings |
| `compare_websites` | Side-by-side scorecard of two sites with a winner |
| `ai_visibility_check` | How 5 major AI engines describe any business + how to improve |
| `get_city_dashboard` | Local economy dashboard for any US city's small business web quality |
| `generate_redesign_preview` | Trigger a free personalized site preview emailed to a user |
All free. All call fivedaylaunch.com's public API. Zero LLM tokens consumed.
## Install (Claude Desktop)
1. Open Claude Desktop → Settings → Developer → "Edit Config" → opens `claude_desktop_config.json`
2. Add:
```json
{
"mcpServers": {
"fivedaylaunch": {
"command": "npx",
"args": ["-y", "github:Nareshdevelop/fdl-mcp-server"]
}
}
}
```
3. Restart Claude Desktop.
4. Look for the 🔌 icon in your chat — fivedaylaunch tools should be listed.
## Example prompts
After installing, ask Claude things like:
- *"Use fivedaylaunch to audit stripe.com"*
- *"Compare nytimes.com and washingtonpost.com using fivedaylaunch"*
- *"Run an AI visibility check on Tesla using fivedaylaunch"*
- *"Show me the fivedaylaunch dashboard for Austin, Texas"*
- *"Use fivedaylaunch to generate a redesign preview of mybusiness.com — email it to me@example.com"*
## Manual install (without Claude Desktop)
```bash
npx -y github:Nareshdevelop/fdl-mcp-server
```
The server runs on stdio. Wire it into any MCP-compatible host (Cursor, Zed, Cline, Continue, etc.).
## What the API returns
Everything is JSON. Public + free.
```bash
# Free public API endpoint — try it yourself:
curl https://fivedaylaunch.com/api/audit?url=stripe.com
# Returns:
{
"domain": "stripe.com",
"score": 85,
"grade": "B",
"breakdown": { ... },
"findings": [...]
}
```
## Why does this exist
We built fivedaylaunch.com as a free public registry of small business website quality. The audit logic is open source ([fdl-site-audit](https://github.com/fivedaylaunch/fdl-site-audit)). This MCP server is the canonical way to access it from inside AI assistants.
We monetize by offering full site rebuilds at **$799 flat, 5 days** — the audit tool is and stays free forever.
## License
MIT © 2026 fivedaylaunch.com
## Related
- **fivedaylaunch.com** — the main product (AI-built websites in 5 days, $799)
- **[fdl-site-audit](https://github.com/fivedaylaunch/fdl-site-audit)** — the deterministic audit core, open source
- **[fivedaylaunch.com/sites](https://fivedaylaunch.com/sites)** — public audit registry
- **[fivedaylaunch.com/ai](https://fivedaylaunch.com/ai)** — AI visibility audit registry
TDQS
Scored across 5 tools
Each tool serves a clearly distinct purpose: auditing a single site, comparing two sites, checking AI visibility, viewing city stats, and generating redesign previews. There is no overlap or ambiguity between them, making it easy for an agent to select the right tool.
Four of five tools follow a consistent verb_noun pattern (audit_website, compare_websites, get_city_dashboard, generate_redesign_preview). 'ai_visibility_check' deviates from the verb-first convention, making it slightly inconsistent but still readable and predictable.
With exactly 5 tools, the set is well-scoped for a website audit/analysis service. Each tool earns its place without redundancy or bloat, and the count is within the ideal 3-15 range.
The tool surface covers the core lifecycle of website analysis: auditing, comparing, AI visibility, city-level aggregation, and redesign previews. Minor gaps exist (e.g., no tool to fetch historical audits or manage saved reports), but the core workflows are well covered.