x=hub

your toolset

    build deployment
    sign in

    openharness

    projectother official scan blocked
    io.github.autonomous-ai/openharness · v
    deploy just this

    The ultimate harness for coding agents and beyond. All your agents. All your machines. One command center. Start with code, then follow your curiosity and build across disciplines: CAD, circuits, robots, games and music.

    run it

    blocked by the security scan
    Blocked by the security scanThe worker scanned github.com/autonomous-ai/openharness at 0c20f1062b5e for v with [email protected][email protected] on 2026-10-17, before building it, and found code that looks malicious: threat-filesystem-autostart in cli/e2e/cli.e2e.ts:155, threat-filesystem-autostart in cli/e2e/harness/daemon.ts:121, threat-network-reverse-shell in companions/src/shared/guard.spec.ts:86, threat-runtime-obfuscation-steganography in devices/harness-device/firmware/scripts/fo_pet.py:281, threat-setup-network-in-install in os/setup.py:529. It built nothing, and the hub offers no way to run this version: no package, no image, no build, no web service. An operator can look at the findings and let it run; a new version is scanned again.

    connect to it

    copy what your client needs
    {
  "mcpServers": {
    "openharness": {
      "command": "CHANGE_ME",
      "args": []
    }
  }
}
    #!/usr/bin/env bash
# Registers every server with Claude Code (run once; `claude mcp list` to check).
set -euo pipefail
set -a; [ -f .env ] && source .env; set +a
claude mcp add openharness -- CHANGE_ME

    {
  "mcpServers": {
    "openharness": {
      "command": "CHANGE_ME",
      "args": []
    }
  }
}
    {
  "servers": {
    "openharness": {
      "type": "stdio",
      "command": "CHANGE_ME",
      "args": []
    }
  }
}
    name: my-agent
target: local
servers:
- hub: io.github.autonomous-ai/openharness
    • not probed yet: tools unpinned, ingest lists them
    • blocked: the worker's security scan of github.com/autonomous-ai/openharness at 0c20f1062b5e on v found code that looks malicious (threat-filesystem-autostart, threat-network-reverse-shell, threat-runtime-obfuscation-steganography, threat-setup-network-in-install); nothing is deployed for it

    tools

    Probe result: faileddocker: security scan: github.com/autonomous-ai/openharness at 0c20f1062b5e has code that looks malicious: threat-filesystem-autostart in cli/e2e/cli.e2e.ts:155, threat-filesystem-autostart in cli/e2e/harness/daemon.ts:121, threat-network-reverse-shell in companions/src/shared/guard.spec.ts:86, threat-runtime-obfuscation-steganography in devices/harness-device/firmware/scripts/fo_pet.py:281, threat-setup-network-in-install in os/setup.py:529

    runs · 1

    the latest tests of this entry by the hub and its workers, newest first
    dateworkermethodstatussandboxreason
    2026-10-17 12:34 local-mac-vm docker failed gvisor docker: security scan: github.com/autonomous-ai/openharness at 0c20f1062b5e has code that looks malicious: threat-filesystem-autostart in cli/e2e/cli.e2e.ts:155…