Open Source · MIT License

textrawl

Personal knowledge base for AI. Crawl your documents, search with meaning.

byJeff Green
Get StartedMCP Tools →

Why textrawl?

Crawl Everything

Markdown, PDF, HTML, email, DOCX. Your documents, structured for AI.

Semantic Search

Find by meaning, not just keywords. Hybrid search fuses keyword and vector results.

Persistent Memory

Remember facts, track conversations, and build a knowledge graph across sessions.

Proactive Insights

Automatically discover connections, patterns, and outliers across your knowledge base.

MCP Native

Built for Claude and other MCP clients. 22 tools for your knowledge in every chat.

Local First

Run entirely on your machine. Your data never leaves.

MCP Tools

Twenty-two tools that give your AI access to everything you know.

Document Tools

search_knowledge

Hybrid semantic + full-text search

search_with_context

Search documents, memories, and conversations

get_document

Retrieve full document content

list_documents

Browse with pagination and filters

update_document

Update document metadata

add_note

Create notes with auto-embedding

knowledge_stats

Knowledge base statistics

Memory Tools

remember_fact

Store facts about entities

recall_memories

Search stored memories

relate_entities

Create entity relationships

get_entity_context

Get all info about an entity

list_entities

Browse known entities

forget_entity

Delete entity and memories

memory_stats

View memory statistics

extract_memories

Extract entities from text via LLM

Conversation Tools

save_conversation_context

Save conversation for recall

recall_conversation

Search past conversations

list_conversations

Browse conversation history

get_conversation

Get full conversation transcript

delete_conversation

Delete a conversation session

conversation_stats

Conversation storage statistics

Insight Tools

get_insights

View discovered patterns and connections

discover_connections

Trigger insight scan

dismiss_insight

Dismiss an insight

insight_stats

Insight queue statistics

Up and running in 60 seconds

1git clone https://github.com/jeffgreendesign/textrawl.git

Clone the repo

2pnpm run setup

Configure credentials

3pnpm run dev

Start the MCP server

Why I built this

I kept losing context. Notes in one app, bookmarks in another, PDFs scattered across folders. When I needed to find something, search always failed me.

Textrawl started as a personal knowledge base—a way to crawl my own documents and make them searchable with semantic understanding, not just keyword matching.

Now it's an MCP server that gives AI assistants access to your knowledge. Your second brain, available in every conversation.

JG
Built byJeff GreenProduct Engineer