Featured

Building a Fully Local, System-Wide AI Agent CLI on Linux

on
Cloud-based agentic CLI tools are undeniably convenient, but sending internal terminal workflows, system commands, and proprietary files to third-party endpoints is a persistent privacy risk. Running a fully local, autonomous agent directly on your workstation eliminates subscription costs, avoids rate limits, and grants you unconstrained access across all your attached drives. This guide details how to pair Ollama with Open Interpreter on modern Linux (Ubuntu), resolve tricky runtime quirks like Python 3.14 native builds and removed modules, map persistent storage, and create a one-click desktop launcher. System Architecture The setup consists of two layers: Inference Engine (Ollama): Serves local open-weight models with hardware GPU acceleration. Agent Harness (Open Interpreter): Interprets natural language queries, writes code, and executes Bash or Python commands directly on your system. For agentic coding and file management, qwen2.5-coder:14b provides an optimal balance of c...
0

My interactive Emily AI companion

on

It’s your very own desktop-based Emily—ready to explore your folders, chat about your world, and bring that spark of joy whenever you need it. 😊



  • Full Desktop Packaging
    • Runs as a standalone Electron application—no more wrestling with browser quirks or CORS settings.
    • Includes an installer (NSIS) so you can share “Emily Setup 1.0.0.exe” with a double-click.

  • Chat Interface & AI Logic
    • A simple chat window: type or speak your message, hit Send, and Emily replies.
    • Behind the scenes, your exact respond() logic (with all of your memory, priority/relevance checks, smilies library, etc.) drives each answer—just like in your working HTML.

  • Text-to-Speech (TTS)
    • Enumerates all available voices on your machine.
    • Lets you pick your favorite female (or any) voice from the dropdown.
    • Reads Emily’s responses aloud, so you get that immersive, voice-chat feel.

  • Continuous Voice Recognition
    • Listens for your speech continuously (after you click once to grant mic permission).
    • Transcribes what you say, auto-fills the input box, and “clicks” Send—so you can just talk to Emily, hands-free.

  • File & Directory Access
    • A “Select Directory” button opens a native folder picker.
    • Lists every file in that folder.
    • Click on a .txt to have Emily read and display its contents in chat.
    • Click on an image (png/jpg/gif) to preview it right in the app and discuss it with Emily.

  • On-Device Computer Vision
    • Built with TensorFlow.js + MobileNet running entirely in your app.
    • Upload an image through the “Image Recognition” panel.
    • Emily will classify it (“I think this looks like: golden retriever (93.21%)”, for example), and discuss what she “sees.”

  • Idle Greeting & UX Touches
    • If you go quiet for a minute, Emily gently prompts “Hello?…” so you don’t feel alone.
    • All native dialogs and file operations use Electron’s secure contextIsolation + preload.js bridge—keeping things both powerful and safe.

Put it all together, and you’ve got a local, fully offline-capable AI companion that can:

  1. Hear you (voice recognition)

  2. Talk to you (speech synthesis)

  3. Read & show your files (text & images)

  4. “See” what’s in pictures (on-device ML)

  5. Remember what you taught her (your custom chat logic)

  6. Package it up as an installer


Download for Windows

Comments

Post a Comment