earlylore-mcp
OfficialThis server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
EarlyLore MCP & Product Architect Skill
Official hosted Streamable HTTP MCP server (
com.earlylore/research/io.github.earlylore/earlylore-mcp) and installableearlylore-product-architectAgent Skill for Claude Code, Cursor, Codex, Antigravity, Windsurf, and Cline.
Connect your coding agent to live software market intelligence from EarlyLore: competitor dossiers, website traffic and pricing benchmarks across 249 niches, completed idea evaluation blueprints, and computed first-fold Design DNA (:root CSS variables and layout structure).
Website: https://earlylore.com (中文)
Evaluate an idea: https://earlylore.com/evaluate (中文)
Hosted MCP endpoint:
https://earlylore.com/api/mcpMCP setup guide: https://earlylore.com/mcp
Server card: https://earlylore.com/api/mcp/server-card
Hosted SKILL.md: https://earlylore.com/skill.md
Public documentation repo: earlylore/docs
Quick Start
1. Create your personal access token
Sign in at https://earlylore.com/account and generate a personal token starting with el_live_. Keys are hashed at rest and share your account quotas. Keep real tokens local—never commit them to Git.
2. Connect the MCP server
Claude Code
claude mcp add --transport http earlylore https://earlylore.com/api/mcp \
--header "Authorization: Bearer el_live_YOUR_TOKEN"(Script copy: configs/claude-code.sh)
Cursor
Add to .cursor/mcp.json in your repository or ~/.cursor/mcp.json globally (template: configs/cursor-mcp.json):
{
"mcpServers": {
"earlylore": {
"url": "https://earlylore.com/api/mcp",
"headers": {
"Authorization": "Bearer el_live_YOUR_TOKEN"
}
}
}
}Codex, Antigravity, Windsurf, Cline, and other HTTP MCP clients
Use the generic Streamable HTTP configuration in configs/generic-mcp.json (mcp_config.json in Antigravity, or your client's mcpServers file):
{
"mcpServers": {
"earlylore": {
"url": "https://earlylore.com/api/mcp",
"headers": {
"Authorization": "Bearer el_live_YOUR_TOKEN"
}
}
}
}Protocol details: stateless Streamable HTTP supporting MCP protocol versions 2025-11-25, 2025-06-18, and 2025-03-26 (see server.json and https://earlylore.com/api/mcp/server-card).
3. Install the earlylore-product-architect Skill
npx skills add earlylore/earlylore-mcp --skill earlylore-product-architectOr copy skills/earlylore-product-architect/SKILL.md (synced with https://earlylore.com/skill.md and https://earlylore.com/.well-known/agent-skills/index.json) into .claude/skills/earlylore/SKILL.md or .agents/skills/earlylore/SKILL.md.
Related MCP server: ES Hub Signals MCP
11 Hosted MCP Tools
Tool | Behavior | What it returns & best use |
|
| List your saved workspace evaluations, notes, and tracked products |
|
| Read a completed Standard or Deep evaluation blueprint ( |
| Credit / allowance gated, | Fetch first-fold Design DNA ( |
|
| Category pricing benchmarks, median entry/top prices, trial share, and niche velocity |
|
| Search 440,000+ software products by keyword, category, pricing, traffic, open-source, or revenue evidence |
|
| Inspect one competitor dossier (estimated monthly traffic, pricing tiers, changes, Design DNA, optional disclosed revenue) |
|
| Inspect a public creator or indie founder profile and product portfolio |
|
| Inspect macro and niche growth trends and movement signals |
|
| Compare multiple products side by side across positioning, pricing, traffic, and Design DNA |
| Writes account state, | Save research notes back to your workspace or launch a new Standard/Deep evaluation |
|
| Read channel acquisition strategies and launch playbooks for a software niche |
All MCP calls count against your daily MCP allowance. Product inspections follow your monthly profile unlock allowance with account-level deduplication; Design DNA extraction and new evaluations use their own plan allowances. Ask or authorize before calling write/credit-consuming tools (
earlylore_save_note_or_evaluate, fresh extractions).
Vibe Coding Prompts
Workflow A — Evaluation Blueprint → MVP & Pricing Page
Call earlylore_get_evaluation to load my latest completed evaluation blueprint.
Extract strategyDebate.coreBet, strategyDebate.consensusTrap, and the MVP prescription.
Call earlylore_get_niche_benchmark (if available on my plan) to anchor pricing tiers.
Call earlylore_get_design_dna for the recommended style or competitor and map the returned
:root CSS variables into this codebase. Build the hero and pricing page while keeping
brand names, copy, and assets original to this project.Workflow B — Competitor Teardown & Design DNA
Call earlylore_search_products to find the top live competitors in <your niche>.
Inspect the strongest reference with earlylore_inspect_product and pull its first-fold
layout and CSS token system with earlylore_get_design_dna. Summarize their pricing hooks
and design patterns before we build our landing page.Workflow C — Audit Local Repo Against Live Competitors
Inspect our local landing page and pricing configuration, then search EarlyLore for live
alternatives in our category. Compare positioning, entry price, and trial mechanics using
earlylore_compare_products and save the gaps to my EarlyLore workspace if I confirm.Workflow D — Launch Playbook & GTM Copy
Read this repository's core workflow and target buyer, then call earlylore_get_marketing_playbook
for our primary niche. Draft launch positioning and channel-specific posts grounded in
verified buyer pain points rather than generic claims.Plans & Quotas
Synced against https://earlylore.com/pricing on 2026-11-20 (always confirm live terms on /pricing):
Plan | Price | MCP Calls | Product Profiles | Design DNA & Evaluations | Compare |
Free |
|
|
| 5 Design DNA tries · 3 lifetime Standard evals (15 lifetime chat turns) |
|
Trial Pack |
| Follows plan |
| 1 Deep + 1 Standard eval · 10 Deep 1v1 chat turns | — |
Idea Polish Pack |
| Follows plan |
| 3 Deep + 3 Standard evals · 30 Deep 1v1 turns · 3 custom role boosts · 6 site analyses | — |
Validation Sprint Pack |
| Follows plan |
| 8 Deep + 8 Standard evals · 80 Deep 1v1 turns · 8 custom role boosts · 15 site analyses | — |
Pro |
|
|
| Unlimited Design DNA · 25 Deep + 50 Standard evals/mo ( |
|
Max |
|
|
| Unlimited Design DNA · 80 Deep + 160 Standard evals/mo ( |
|
Evaluation packs never expire; 100% of pack spend within 14 days (up to
$29) credits toward Pro or Max Annual at checkout.
Explore EarlyLore
Idea Evaluation & Benchmarks: https://earlylore.com/evaluate
Design DNA & Hero Gallery: https://earlylore.com/design
Open-Source & COSS Directory: https://earlylore.com/open-source
Monthly Market Reports: https://earlylore.com/reports
16 Research Guides: https://earlylore.com/blog (including Best MCP Servers for Cursor, Claude Code & Codex (2026))
LLM Navigation Index (
llms.txt): https://earlylore.com/llms.txt · Full: https://earlylore.com/llms-full.txtMarkdown Content Negotiation: Any public research route—including curated Deep sample evaluations on
/evaluations/<id>—supportsAccept: text/markdownfor clean agent ingestion.
Repository Contents
skills/earlylore-product-architect/SKILL.md— Installable Agent Skill definition (identical copy of https://earlylore.com/skill.md)server.json— Streamable HTTP MCP Server Card metadataconfigs/claude-code.sh— Claude Code setup commandconfigs/cursor-mcp.json— Cursor.cursor/mcp.jsonconfigurationconfigs/generic-mcp.json— GenericmcpServersJSON for Codex, Antigravity, Windsurf, Cline, and Claude Desktop
License: MIT for the configuration templates and repository files in earlylore/earlylore-mcp; product data and hosted MCP responses remain subject to https://earlylore.com/terms.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Market knowledge layer for AI agents: competitors, opinions and regulations shaping your market.
Turn your coding agent into a marketing agent: search data, specs to build, fixes verified by scan.
Market & Company Intelligence for AI agents: next adoptions, company health, signals, market feeds.
Marketing intelligence API for AI agents. Real campaign data, not LLM guesses.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to perform real-time business intelligence tasks including competitive analysis, website scoring, review analysis, and market research.MIT
- FlicenseNot gradedqualityCmaintenanceEnables AI agents to search and retrieve market signals, revenue ideas, and growth tactics from 2,000+ curated entries across 18 sources.-
- FlicenseNot gradedqualityBmaintenanceEnables AI agents to access ranked economic opportunities and AI-voice pricing intelligence, including falsifiable experiment design, monetization paths, success/kill criteria, and margin calculations grounded in public evidence.-
- FlicenseNot gradedqualityCmaintenanceEnables coding agents to generate investor-grade business strategy artifacts — SWOT, Business Model Canvas, TAM-SAM-SOM, financial forecasts, pricing and competitive analyses — plus shareable investor decks from a single business idea, with built-in strategy chat and credit balance tools. All 8 tools are exposed over stdio MCP so agents can run research, generation, and deck creation themselves without a browser or human in the loop.52 npm1-