google/sec-gemini

Sec-Gemini is a cutting-edge AI model designed to enhance cybersecurity capabilities and empower defenders in the ongoing battle against cyber threats.

What is Sec‑Gemini?

Sec‑Gemini is an experimental AI system built by Google for cybersecurity use‑cases. This repository provides software development kits (SDKs) that let developers call the model from their own code or embed a chat widget on a web page.

Available SDKs

Language Location
Python sec-gemini-python/
TypeScript sec-gemini-ts/

Both SDKs expose the same underlying API, allowing you to send prompts to Sec‑Gemini and receive its responses.

Web component

A ready‑to‑use web component (<sec-gem-chat>) is supplied for quick integration into any site. Example usage:

<sec-gem-chat
  incognito="true"
  session-id=""
  session-name="TestName"
  session-description="TestDescription"
  session-prompt=""
  theme="dark"
  api-key="YOUR_API_KEY">
</sec-gem-chat>
<script src="https://cdn.jsdelivr.net/npm/sec-gemini-web-component/dist/swc.iife.js"></script>

The component handles the UI and communicates with the Sec‑Gemini backend using the provided API key.

Where to learn more

The project points to a dedicated website: https://secgemini.google. That site contains further documentation, pricing, and contact information.

Disclaimer

  • This is not an officially supported Google product.
  • The repository is ineligible for Google’s Open Source Software Vulnerability Rewards Program.

Bottom line: If you need a programmable interface to Google’s experimental cybersecurity AI—whether from Python, TypeScript, or a browser widget—this repo gives you the starter kits. It’s a thin wrapper around the service rather than a full‑stack security platform.

Related

  • Project
    johnmarktaylor91/torchlens

    TorchLens is a Python library for capturing, visualizing, and intervening on the full computational graph of any PyTorch model (and preview support for other frameworks). It records every activation and gradient, provides rich per‑operation metadata, lets you query or filter tensors, draw PDF graphs, compute receptive/projective fields, and replay or modify the graph for “what‑if” experiments. The tool is validated on >11 600 architectures, with ~89 % algorithmically verified for faithful capture.

  • Project
    Varnan-Tech/opendirectory

    OpenDirectory is a catalog of 64 plug‑in‑style “agent skills” that add marketing, growth and visual‑content expertise to LLM agents (Claude, Gemini, Codex, etc.). Install a skill with a single `npx` command (or via skills.sh, Claude Desktop, or Manus AI) and the target agent instantly gains the prompts, context and tool definitions needed for tasks like blog‑cover generation, brand naming, pricing‑page audits, competitor intel, and video creation. It’s a reusable, versioned prompt‑library aimed at founders and marketers who want AI assistance without writing prompts from scratch.

  • Project
    visualbruno/3DGenStudio

    3D Gen Studio is an open‑source, AI‑driven desktop/web app that lets you build 3‑D assets end‑to‑end. It combines a Kanban board or node‑graph UI with ComfyUI workflows and external 3‑D generation APIs, offering image‑to‑mesh pipelines, auto‑UV/retopo, AI‑auto‑rigging (SkinTokens + optional Kimodo motion), LOD/optimization, and game‑ready checks—all stored locally on disk.

  • Project
    open-mmlab/FoleyCrafter

    FoleyCrafter is a video-to-audio generation framework that produces realistic, synchronized sound effects based on the visual content of a silent video.