RAG Frameworks · lucasastorian

llmwiki

LLM Wiki is an open-source Python application that turns your documents, notes, and web clips into an automatically maintained personal knowledge base. Claude reads your sources via an MCP connection and periodically synthesizes them into a cross-linked wiki that grows over time without manual curation.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/lucasastorian/llmwiki

1.3k

GitHub stars

205

Forks

Python

Primary language

Apache-2.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	lucasastorian/llmwiki
Owner	lucasastorian
Primary language	Python
License	Apache-2.0 — OSI-approved
Stars	1.3k
Forks	205
Open issues	17
Latest release	Unknown
Last updated	2026-07-08
Source	https://github.com/lucasastorian/llmwiki

What llmwiki is

Built on Python 3.11+, Node.js 20+, Next.js frontend, and SQLite local indexing, LLM Wiki integrates Claude via MCP for autonomous wiki generation and maintenance. It supports multi-format ingestion (PDF, Office, Markdown, HTML), file watching, and scheduled Claude Routines for incremental updates.

Quickstart

Get the llmwiki source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/lucasastorian/llmwiki.git2cd llmwiki3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Personal Research & Learning Compounding

Researchers, academics, and lifelong learners who want a persistent record of what they read and thought about it—with automatic synthesis into interconnected notes—without manual upkeep.

Organizational Knowledge Base with Minimal Overhead

Teams seeking to build institutional memory from scattered documents and notes without requiring dedicated knowledge management staff; Claude refreshes the wiki nightly as new sources arrive.

AI-Enhanced Context Layer for Agent Workflows

Developers building Claude-based agents who need a searchable, structured knowledge base that the AI can read, update, and cross-reference to apply consistent mental models.

Implementation considerations

Local file watching and indexing work out of the box, but multi-format support (Office, PDFs) requires optional external tools: LibreOffice for Word/PowerPoint, and Mistral API key for OCR quality.
Chrome extension and local workspace sync are straightforward, but deploying to remote/cloud requires additional Supabase or database setup not fully detailed in README.
Claude Routine scheduling works via Claude.ai/Code or Desktop scheduled tasks; timing and resource costs depend on workspace size and Claude API pricing per run.
MCP connection requires manual JSON config copy-paste into Claude Desktop or Claude Code; no automated onboarding or role-based access control is evident.
Wiki pages are ordinary Markdown files stored on disk; version control and concurrent edits are possible but conflict resolution between human edits and Claude's updates is not explicitly addressed.

When to avoid it — and what to weigh

Real-Time, Mission-Critical Knowledge Systems — If you need sub-second query responses, ACID guarantees, or real-time consistency across many concurrent users, the local SQLite + nightly batch approach will not meet those SLAs.
Strict Data Governance or Compliance Requirements — Requires careful vetting before use in regulated environments (healthcare, finance) because Claude processes your source documents over MCP; data residency and audit trails must be evaluated.
Large Multi-Tenant or Enterprise Deployments — The architecture (one workspace = one MCP server) does not cleanly scale to hundreds of isolated knowledge bases; requires custom infrastructure work to isolate tenants.
Fully Offline or Air-Gapped Operation — Claude Routines and MCP require external API calls to Anthropic; local indexing works offline, but the autonomous wiki-building feature requires internet connectivity.

License & commercial use

Apache License 2.0 (Apache-2.0). This is a permissive OSI-approved license that allows commercial use, modification, and redistribution with minimal restrictions (requires license and copyright notice, no warranty/liability).

Apache 2.0 explicitly permits commercial use. You may deploy this internally, build products on top of it, or run it as a service—provided you retain license notices. However, any commercial deployment should independently verify data handling, compliance, and liability assumptions, especially because it integrates Claude API calls.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	High

Security considerations

LLM Wiki processes your documents through Claude API via MCP, so all source material transits Anthropic's infrastructure. Local indexing (SQLite) is not encrypted at rest. No authentication or access control is built in; the app assumes single-user or trusted-network use. Chrome extension communication is unencrypted over localhost. No audit logging or data retention policies are mentioned. Before production or regulated use, review data handling practices, API security, and whether Anthropic's terms permit your use case.

Alternatives to consider

Obsidian + Obsidian Copilot

Offline-first note-taking with optional AI integration; better for local-only workflows but less autonomous wiki synthesis and no built-in multi-format ingestion.

Notion + Custom Claude Integration

Notion provides rich collaborative docs and databases; integrating Claude manually via API gives you control but requires custom build-out and lacks the turnkey MCP + file-watching architecture.

Software development agency

Build on llmwiki with DEV.co software developers

Clone LLM Wiki, connect Claude via MCP, and start ingesting your research. Or discuss custom deployment, scaling, and integration with our team.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

WikiChat llm-for-zotero llm-graph-builder

Related on DEV.co

Explore the category and the services that help you build with it.

RAG Frameworks AI Application Development Web App Development API Development Talk to DEV.co

llmwiki FAQ

Can I use LLM Wiki offline?

Local indexing and file serving work offline. Claude Routines and MCP require API calls to Claude, so autonomous wiki maintenance and AI integration require internet. Files stay on your machine.

Does LLM Wiki upload my files to the cloud?

No. Source files stay where you point it (e.g., ~/research). Local mode generates wiki/ and .llmwiki/ in that folder. Remote mode and hosted version behavior is not fully detailed; verify before using llmwiki.app with sensitive data.

What if Claude and I both edit the same wiki page?

Not explicitly covered in README. Pages are ordinary Markdown; concurrent edits could conflict. The guide suggests Claude appends to wiki/log.md, but merge strategy between human edits and AI updates is unclear.

How much does it cost to run?

Self-hosting is free (open source, MIT/Apache license covers code). Claude API calls per routine run incur per-token charges (Anthropic pricing); at scale, nightly refreshes across large workspaces could be non-trivial. Not quantified in README.

Software developers & web developers for hire

Need help beyond evaluating llmwiki? DEV.co is a software development agency offering software development services and web development for teams of every size. Our software developers and web developers build custom software, web applications, APIs, and rag frameworks integrations — and maintain them long-term.

Ready to Build an AI-Maintained Knowledge Base?

Clone LLM Wiki, connect Claude via MCP, and start ingesting your research. Or discuss custom deployment, scaling, and integration with our team.

Talk to DEV.co Schedule a Consultation