Skip to content
View PeepSick's full-sized avatar

Block or report PeepSick

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
PeepSick/README.md
Peepsick Labs — Technology and AI Ecosystem

Peepsick Labs

We build the infrastructure. Agents do the work.

AI Infrastructure · Agent Systems · Developer Tools · Autonomous Software

PEEPSICK LIMITED · London, United Kingdom · Independently funded · Building in public


Website · GitHub · Contact


I. About Us

Peepsick Labs is an independent software company focused on AI infrastructure, agent systems, developer tools and autonomous software.

We design and build practical systems that help AI agents discover capabilities, interact with software, work across environments and get things done.

Our work spans open-source infrastructure, independent software products and connected AI ecosystems.

We believe in useful engineering, portable systems and products that solve real problems.


II. Our Ecosystem

A growing collection of products and technologies, each designed to solve a different part of the AI infrastructure puzzle.

Project Description Status
🛠️ USB+ — Universal Systems Bridge An open-source bridge for discovering, installing and verifying AI skills across agent runtimes. Includes secscan, mcprobe, SHA-256 integrity verification, lock, diff and repair. 529 skills and 16 runtime targets. 🟢 Live · v1.2.0
🔎 IzMercek AI Visibility Intelligence Platform. Helping businesses understand how their websites and brands appear across AI-powered search and answer experiences. 🟢 Live
🧠 LeoSIS Ecosystem An AI agent ecosystem built around the LeoSIS runtime, autonomous workflows, multi-agent orchestration, persistent context and integrations. 🔵 Coming soon
🎙️ ModelForce Self-hosted voice infrastructure with a unified interface for multiple TTS providers, including Piper, Kokoro and XTTS. 🛠️ Active development
📬 Keep E Box An agent-native, provider-agnostic mail workspace connecting major mail providers through a unified model, agent layer, policy engine and SDK. 🧪 Internal development
🏭 FinishOps Release operations console for repository auditing, release readiness, Definition of Done, technical debt and release workflows. 🧪 Internal development
Peepsick Labs ecosystem

USB+ connects capabilities. IzMercek measures AI visibility. LeoSIS powers intelligent agents. ModelForce gives AI a voice. Keep E Box connects agents to communication. FinishOps helps teams ship.


III. USB+ — Universal Systems Bridge

One bridge for AI capabilities, runtimes and trust.

USB+ started with a simple idea: AI skills and capabilities should be portable across runtimes, easy to discover and easier to verify.

It has evolved into an open-source ecosystem combining capability discovery, installation and a dedicated trust layer.

The Trust Layer:

  • secscan — Static skill contract verification with file and line-level evidence.
  • mcprobe — Consent-based runtime observation of MCP servers.
  • verify — SHA-256 integrity checks for installed skills.
  • lock — Reproducible, version-pinned installations.
  • diff — Capability and permission change inspection.
  • repair — Restore verified installed files.
  • GitHub Code Scanning support through SARIF output.

Current release: v1.2.0

License: MIT

Explore USB+ · GitHub Repository · npm CLI · npm SDK


IV. LeoSIS — Agent Ecosystem

LeoSIS is part of the core Peepsick technology ecosystem.

It is not just another chatbot. The project explores autonomous agent execution, computer interaction, multi-agent collaboration, persistent context and connected tools.

The LeoSIS ecosystem is designed around modular capabilities and a runtime that can bring different models, tools and workflows together.

Its development includes:

  • Autonomous agent runtime
  • Computer interaction and automation
  • Multi-agent orchestration
  • Persistent sessions and context
  • Voice interaction
  • Tool and MCP integrations
  • Business and workflow intelligence

Public release and product details will be announced separately.


V. How We Work

  • Build in public. We share useful tools, experiments and selected parts of our development process.
  • Honest engineering. Released products, active development and upcoming projects are clearly distinguished.
  • Open source where it makes sense. We publish useful infrastructure and developer tools whenever practical.
  • Portable by design. We avoid unnecessary dependence on a single provider, framework or runtime.
  • Small teams, real products. We focus on shipping useful systems rather than adding complexity for its own sake.

VI. Technology


VII. GitHub Activity

Peepsick GitHub Stats

VIII. Open Source & Collaboration

We enjoy working with people who build things, challenge assumptions and contribute practical ideas.

Bug reports, documentation improvements, new runtime adapters, security research and meaningful pull requests are welcome.

Explore our repositories and build something with us.

Peepsick Labs Workshop



Built with curiosity at Peepsick Labs.

Independent software. Connected intelligence. Useful technology.


🌐 peepsicklabs.com · 🛠️ USB+ · 🔎 IzMercek · ✉️ info@peepsicklabs.com

PEEPSICK LIMITED · London, United Kingdom

Pinned Loading

  1. ModelForce ModelForce Public

    A runtime and provider infrastructure for self-hosted text-to-speech.

    TypeScript

  2. usb usb Public

    USB - Universal Skill Bridge: one portable skill format, installed into 16 agent runtimes with sha256-verified installers

    TypeScript 7 1

  3. STELLORA STELLORA Public

    Every memory is a star. Every story is a constellation.

    TypeScript 1

  4. usbplus usbplus Public

    USB+ — Universal Systems Bridge. One skill format, every agent runtime: 529 skills, 16 target formats, Trust Layer (secscan + mcprobe), zero-dependency CLI, HTTP MCP server.

    Python 12 2