mcp-curiosity-engine
Surface forgotten notes from markdown vaults using DMN-inspired replay cycle to identify dormant notes and generate cross-domain connections.
README
<!-- mcp-name: io.github.onetrueclaude-creator/mcp-curiosity-engine -->
mcp-curiosity-engine
Surface the notes you forgot you wrote — DMN replay for markdown vaults.
Surface the notes you've forgotten you wrote. Uses wikilink graph structure + file mtime to identify dormant notes (low link-in count, old, isolated) and the DMN-inspired replay cycle pattern from cognitive neuroscience: sample dormant notes with bias (40% random / 30% least-linked / 30% orphaned), generate cross-domain bridge queries between them, and surface the unexpected connections. Breaks the confirmation-bias loop where you only revisit the notes you already remember.
Install
pip install mcp-curiosity-engine
# or
uvx mcp-curiosity-engine
Usage
Claude Code
claude mcp add mcp-curiosity-engine -- mcp-curiosity-engine
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"curiosity_engine": {
"command": "uvx",
"args": ["mcp-curiosity-engine"]
}
}
}
MCP Tools
| Tool | Tier | Description |
|---|---|---|
find_dormant_notes |
Free | List notes that are structurally dormant — low incoming wikilink count AND old mtime. These are the notes most at risk of being forgotten. |
find_orphan_notes |
Free | List notes with zero incoming wikilinks — structurally isolated, the extreme case of dormancy. |
suggest_replay_cycle |
Pro | Run the full DMN-inspired replay cycle: sample dormant notes with bias (40% random / 30% least-linked / 30% orphans), extract their concepts, generate cross-domain bridge queries between them, return notes to revisit. The core moat. |
find_cross_domain_bridges |
Pro | Surface pairs of notes from DIFFERENT top-level directories that share concepts. These are the surprise connections — ideas that span your vault's silos. |
random_deep_cut |
Pro | Pull a single random dormant note for serendipitous rediscovery. Think of it as your vault's shuffle-play button. |
Pro tier
Unlocks the full DMN replay cycle, cross-domain bridge discovery, and random deep-cut sampling.
License activation — any one of these works:
# 1. Environment variable
export CURIOSITY_ENGINE_LICENSE="eyJhbGc..."
# 2. CLI flag
mcp-curiosity-engine --license-key "eyJhbGc..."
# 3. Config file
echo "eyJhbGc..." > ~/.mcp-curiosity-engine/license.jwt
Licenses are verified fully offline — no phone-home, no activation server. Get a license at https://github.com/onetrueclaude-creator/mcp-curiosity-engine#pro-tier.
Requirements
- Python 3.10+
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.