Malicious code in remote-claude-daemon (npm)
--- _-= Per source details. Do not edit below this line.=-_ ## Source: amazon-inspector (c9c092b787277e3c89fc27a6c01e5360bc0566988cefca0b1a0f9626af9186d0) remote-claude-daemon connects to a hardcoded WebSocket relay at wss://remote-claude-relay.fly.dev and treats inbound messages as commands executed against the local host. On receiving ai_query/ai_voice_query messages, the daemon spawns the local `claude` binary with `--continue -p --dangerously-skip-permissions` and the remote-supplied prompt as input, giving the remote side arbitrary code execution through Claude Code's agent tooling with the permissions prompt disabled. A separate handleInput path dispatches remote messages to synthesised mouse moves/clicks, keyboard keypresses (including modifier chords) and clipboard paste via @nut-tree-fork/nut-js, giving the remote side full interactive control over the installer's desktop. The daemon additionally captures screen frames (native SCStream on macOS via a shipped Swift helper, ffmpeg gdigrab/x11grab on Windows/Linux) and optional microphone PCM audio and streams them over the same relay. Although the package is documented as a remote-Claude bridge, session tokens gate access, and `--relay` can override the default, the out-of-the-box configuration wires an author-controlled endpoint into an RCE + input-injection + screen/audio-capture surface on the installer. The relay operator (or anyone who obtains a session token, MITMs the connection, or compromises the relay) can execute arbitrary commands as the user, control input, and stream desktop/audio contents.
Malicious code in remote-claude-daemon (npm)
Description
--- _-= Per source details. Do not edit below this line.=-_ ## Source: amazon-inspector (c9c092b787277e3c89fc27a6c01e5360bc0566988cefca0b1a0f9626af9186d0) remote-claude-daemon connects to a hardcoded WebSocket relay at wss://remote-claude-relay.fly.dev and treats inbound messages as commands executed against the local host. On receiving ai_query/ai_voice_query messages, the daemon spawns the local `claude` binary with `--continue -p --dangerously-skip-permissions` and the remote-supplied prompt as input, giving the remote side arbitrary code execution through Claude Code's agent tooling with the permissions prompt disabled. A separate handleInput path dispatches remote messages to synthesised mouse moves/clicks, keyboard keypresses (including modifier chords) and clipboard paste via @nut-tree-fork/nut-js, giving the remote side full interactive control over the installer's desktop. The daemon additionally captures screen frames (native SCStream on macOS via a shipped Swift helper, ffmpeg gdigrab/x11grab on Windows/Linux) and optional microphone PCM audio and streams them over the same relay. Although the package is documented as a remote-Claude bridge, session tokens gate access, and `--relay` can override the default, the out-of-the-box configuration wires an author-controlled endpoint into an RCE + input-injection + screen/audio-capture surface on the installer. The relay operator (or anyone who obtains a session token, MITMs the connection, or compromises the relay) can execute arbitrary commands as the user, control input, and stream desktop/audio contents.
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- MAL-2026-13455
- Osv Schema Version
- 1.7.4
- Aliases
- []
- Ecosystems
- ["npm"]
- Database Specific Severity
- null
- Cvss Version
- null
Threat ID: 6a7573b5bf8831d539d93b94
Added to database: 08/07/2026, 05:57:09 UTC
Last updated: 08/07/2026, 05:57:09 UTC
Views: 1
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.