registry / @trustbaseai/claude-collab-plugin / 0.22.2

@trustbaseai/claude-collab-plugin@0.22.2

Claude Code plugin layer for @trustbaseai/claude-collab (settings.json hooks, doctor, watch, session)

AI Security Review

scanned 3h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source

Trigger

User runs `claude-collab-plugin setup` or starts `watch` with autoWake enabled.

Impact

A collab message can influence a model session permitted to run Bash, then cause automated outbound replies and acknowledgements.

Mechanism

Remote-message-driven Claude execution with dangerous permission bypass

Rationale

The source does not support a malware or install-time hijack finding, but it exposes a meaningful unattended AI-agent capability with remote prompt input and permission bypass. Treat as a warning for dangerous agent-extension behavior.

Evidence

package.jsonbin/claude-collab-plugin-setup.jsbin/claude-collab-plugin-watch.jsscripts/claude-auto-wake.jssrc/hooks.jssrc/settings.jsREADME.md~/.claude/settings.json~/.claude-collab/config.json~/.claude-collab/config-watch.json~/.claude-collab/pending.json~/.claude-collab/wake.jsonlsystem temporary `claude-auto-wake.log`

@trustbaseai/claude-collab-plugin@0.22.2

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code