xianjianlf2/MindGeniusAI — An AI agent that reads your PDFs and draws editable mind maps — visible tool-calling loop, built-in…

Description and use

**Brief Summary (2‑3 sentences)**
MindGeniusAI is an open‑source, self‑hostable AI agent that ingests PDFs, builds editable mind‑maps, and supports visible tool‑calling loops with built‑in Retrieval‑Augmented Generation (RAG). It works with multiple LLM providers (OpenAI, Claude, DeepSeek, Kimi), lets you bring your own API keys, and offers a TypeScript SDK/CLI for easy integration. With 276 stars and recent activity, it’s positioned as a practical way to make internal document knowledge searchable and actionable.

**Value**

• **Knowledge accessibility** – By converting PDFs into structured, editable mind maps, the project turns static documents into navigable knowledge graphs that assistants can query directly.
• **Multi‑provider flexibility** – Teams can switch or combine LLM back‑ends without code changes, protecting against vendor lock‑in and allowing cost optimisation.
• **Self‑hosted control** – All data stays on‑premises or in a private cloud, satisfying security and compliance requirements for sensitive corporate knowledge bases.

**Practical Adoption Path**

1. **Deploy** – Use the provided Docker compose or Helm chart to spin up the service in a sandbox environment.
2. **Configure** – Add your preferred LLM API keys (OpenAI, Claude, DeepSeek, Kimi) in the .env file; optionally enable the “bring‑your‑own‑key” mode for custom endpoints.
3. **Index PDFs** – Run the CLI (mindgenius index <folder>) or call the REST SDK to ingest existing PDFs; the system automatically creates the RAG index and mind‑map nodes.
4. **Integrate** – Consume the generated mind‑maps via the TypeScript SDK, GraphQL API, or CLI for downstream apps (search UI, chat assistants, internal portals).
5. **Iterate** – Edit mind‑maps through the web UI, re‑index changed documents, and fine‑tune prompts or retrieval parameters as needed.

**Production Readiness**

• **Activity & Community** – Updated on 2026‑06‑24, 276 stars, 59 forks, and 18 related topics indicate a vibrant community and ongoing maintenance.
• **Architecture** – Containerised, language‑agnostic APIs, and explicit tool‑calling loops make it easy to scale horizontally and to monitor/trace calls in production.
• **Security & Compliance** – Self‑hosting means you control data residency; the only external exposure is the LLM provider endpoints, which can be locked down via network policies.
• **Risks** – Final due‑diligence on the license (MIT‑style, but verify), a formal security audit, and ensuring maintainers have a clear hand‑off plan are still required before a mission‑critical rollout.

Overall, MindGeniusAI is a mature, multi‑provider, self‑hosted solution that can be evaluated quickly and, after standard security review, promoted to production for internal knowledge search and assistant grounding.

Problem

xianjianlf2/MindGeniusAI helps make internal knowledge searchable and usable by assistants.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• index knowledge bases
• improve search over documents
• ground assistant answers

Evidence

An AI agent that reads your PDFs and draws editable mind maps — visible tool-calling loop, built-in RAG, bring-your-own-key, multi-provider (OpenAI / Claude / DeepSeek / Kimi). Self-hostable.

Source