Skip to content
View Kanevry's full-sized avatar
👨‍🍳
👨‍🍳
8 followers · 3 following

Achievements

Achievement: StarstruckAchievement: Pull Sharkx3Achievement: YOLOAchievement: QuickdrawAchievement: Pair Extraordinairex3

Achievements

Achievement: StarstruckAchievement: Pull Sharkx3Achievement: YOLOAchievement: QuickdrawAchievement: Pair Extraordinairex3

Block or report Kanevry

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.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. 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
Kanevry/README.md

Hi, I'm Bernhard 👋

I build agentic AI systems, and I teach people how to build with them. Based in Vienna.

Things I've open-sourced

  • session-orchestrator — turns ad-hoc Claude Code / Codex / Cursor sessions into a repeatable loop: research, plan, execute in waves, close. Verification gates between waves so regressions don't slip through.
  • aegis — a hardening layer for AI agents. Five-layer defense, with safety treated as a first-class error. Built at the Codex Vienna hackathon.
  • wondraiwork — a framework for AI-assisted open-source contribution. The human leads, the AI assists.
  • DevWatchdog — a small macOS menu-bar app that hunts down zombie Node test and build processes.
  • bg-pdf-service — self-hosted HTML-to-PDF in one command, no vendor lock-in.

Courses — agenticbuilders.at

Hands-on courses on building real things with AI agents. The two closest to the work above:

  • Multi-Agent Orchestration — leading several agents in coordinated waves instead of one endless chat. It's the model session-orchestrator automates.
  • Loop Engineering — autonomous loops that finish verifiably: done-conditions, deterministic gates, kill-switches.

Elsewhere — AI consulting at gotzendorfer.at · @_Kanevry

Pinned Loading

  1. bitgn-pac-agent-public bitgn-pac-agent-public Public

    Winning entry for the BitGN Personal Agent Challenge (PAC) 2026 — defense-in-depth personal agent with soft SGR, layered prompt-injection hardening, and Vercel AI SDK native tool calling.

    TypeScript 6 2

  2. session-orchestrator session-orchestrator Public

    Loop engineering for AI coding agents — turn ad-hoc sessions into a repeatable research → plan → wave-execute → close loop with verification gates. Runs on Claude Code, Codex CLI, Cursor, and Pi. M…

    JavaScript 34 3

  3. claude-code-skills claude-code-skills Public

    Production-tested Claude Code rules for multi-session safety. Prevents work loss when parallel agents share a workspace.

    1

  4. wondraiwork wondraiwork Public

    Systematic AI-assisted open source contribution framework. Human leads. AI assists. Open source benefits.

    Shell

  5. DevWatchdog DevWatchdog Public

    macOS menu bar app that monitors and kills zombie Node.js processes (Vitest, Jest, tsc, esbuild)

    Swift 1