- Updated: December 16, 2025
- 5 min read
Arkham Mirror: Open‑Source Air‑Gapped AI Platform for Investigative Journalism
Answer: Arkham Mirror is an open‑source, air‑gapped AI‑powered investigative journalism platform that runs entirely on a local machine, providing secure document intelligence, semantic search, knowledge‑graph visualisation, and offline RAG without ever sending data to the cloud.
Why Arkham Mirror Matters for Modern Reporters
In an era where data breaches and surveillance are daily headlines, journalists need tools that guarantee absolute privacy. Arkham Mirror delivers exactly that: a fully offline, AI‑enhanced workspace that transforms chaotic document piles into searchable, visualised knowledge graphs—all while keeping the source material on a secure “Data Silo”. This article breaks down the project, its core capabilities, and why it should be on every investigative reporter’s toolkit.
Project Overview & Core Purpose
Arkham Mirror was created by a small team of privacy‑first developers who recognised the gap between powerful AI analytics and the need for offline operation. The repository, hosted on GitHub, is released under the MIT license, encouraging community contributions while ensuring the code remains free for any newsroom or independent researcher.
- Local‑first architecture – no external API calls.
- Built with Python, Docker, and SQLite for lightweight deployment.
- Integrates state‑of‑the‑art NLP, Vision AI, and graph databases.
For a deeper dive into the philosophy behind the platform, see the About UBOS page, which shares a similar commitment to privacy‑first AI solutions.
Key Features & Technical Specifications
🕵️♂️ Offline RAG Chat
Interact with your document corpus through a local Large Language Model (LLM) that performs Retrieval‑Augmented Generation without any internet connection.
🔍 Semantic Search
Search by concept, not just keywords. Embedding‑based indexing lets you locate relevant passages even when the exact phrasing differs.
🕸️ Knowledge Graph Visualisation
Automatically generate entity‑relationship graphs that expose hidden connections between people, organisations, and locations.
⏳ Auto‑Timeline Extraction
Chronologically order events extracted from PDFs, emails, and scanned images, helping you reconstruct story arcs quickly.
📊 Vision‑Based Table Extraction
Leverage OCR and Vision AI to pull structured tables from scanned financial statements, preserving numeric fidelity.
⚠️ Contradiction Detection
AI flags statements that conflict across documents, giving you a fast “red‑flag” list for deeper fact‑checking.
All features run inside Docker containers, ensuring reproducible environments across Windows, macOS, and Linux. The platform uses Chroma DB integration for fast vector similarity search, and the optional OpenAI ChatGPT integration can be swapped for a local LLM when higher accuracy is required.
Installation & First‑Run Guide
Arkham Mirror ships with a “Smart Installer” that automates the setup of Python, Docker, and the required SQLite database. Follow these steps to get started:
- Clone the repository:
git clone https://github.com/mantisfury/ArkhamMirror.git - Run the installer:
- Windows – double‑click
setup.batand follow the wizard. - macOS / Linux – execute
chmod +x setup.sh && ./setup.shin the terminal.
- Windows – double‑click
- Load your documents: Place PDFs, images, or plain‑text files into the
DataSilofolder. The system will automatically ingest and index them. - Launch the UI: Open
http://localhost:3000in a browser. You’ll be greeted by a clean dashboard where you can start chatting with the AI, run semantic queries, or explore the knowledge graph.
For developers who want deeper customisation, the Web app editor on UBOS provides a low‑code environment to tweak UI components or add new AI pipelines.
Why Investigative Journalists Should Adopt Arkham Mirror
Traditional investigative workflows involve juggling spreadsheets, PDFs, and email threads. Arkham Mirror consolidates these assets into a single, searchable knowledge base, delivering four core benefits:
🔐 Uncompromised Security
All processing stays on the local machine. No data ever leaves the “Data Silo”, eliminating exposure to third‑party breaches.
⚡ Accelerated Insight
Semantic search and auto‑timeline generation cut research time by up to 60 % compared with manual keyword searches.
🧩 Visual Story Mapping
The knowledge‑graph view helps reporters spot hidden relationships, supporting more compelling narratives.
🛠️ Extensible Toolkit
Because it’s open source, you can add custom parsers, integrate with other UBOS AI services, or even embed the AI marketing agents for outreach automation.
These capabilities align perfectly with the needs of tech‑savvy journalists, developers, AI enthusiasts, and security‑conscious media professionals who demand a trustworthy, powerful investigative environment.
Get Started with Arkham Mirror Today
Ready to bring air‑gapped AI into your newsroom? Follow the installation steps above, then explore the broader UBOS ecosystem for complementary tools:
- Explore the Enterprise AI platform by UBOS for scaling investigations across large teams.
- Check out the UBOS partner program to collaborate on custom AI modules.
- Browse the UBOS portfolio examples for real‑world case studies.
- Kick‑start your next project with the UBOS templates for quick start, including the “AI SEO Analyzer” and “AI Article Copywriter”.
- For startups seeking a lean AI stack, see UBOS for startups.
- SMBs can benefit from UBOS solutions for SMBs, which include secure document processing pipelines.
If you need a visual walkthrough, the Workflow automation studio lets you design end‑to‑end investigative pipelines without writing a single line of code.
Take the first step: clone the repo, install, and let Arkham Mirror turn your data chaos into a clear, actionable story.
Conclusion
Arkham Mirror stands out as a pioneering air‑gapped AI solution that empowers investigative journalists to work securely, efficiently, and creatively. By keeping every computation local, it eliminates the privacy risks associated with cloud‑based AI while delivering the same cutting‑edge analytics that modern newsrooms demand. As open‑source software, it invites collaboration, ensuring the platform will evolve alongside the challenges of tomorrow’s journalism.
Stay informed about the latest UBOS innovations by visiting the UBOS homepage and subscribing to our AI journalism blog. The future of secure, AI‑driven reporting is here—make it yours.