alvin-bot@5.131.0
Self-hosted autonomous AI agent on Telegram, Slack, Discord, WhatsApp, Signal, terminal & web. Claude Agent SDK + multi-provider engine with auto-failover, detached sub-agents, zero-config memory. Local-first, telemetry-free.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiNo install-time malware chain was found, but the enabled plugin has a concrete command-injection flaw and the package contains opaque, permission-bypassing agent execution. This warrants a warning for a critical vulnerability rather than a malware block.
Decision evidence
public snapshot- The published distribution is deliberately obfuscated, making its runtime behavior difficult to audit.
- The Claude bridge renames platform and tool identifiers as a stated detection bypass, then launches Claude with permission checks disabled.
- The email plugin passes chat-controlled search text directly into a shell command, creating command injection when the plugin is enabled.
- The manifest has no preinstall, install, or postinstall hook; its only lifecycle hook is prepublishOnly.
- There is no dependency on another release of this package.
- The inspected bridge and email sources do not show automatic secret exfiltration or a maintainer-controlled network endpoint.
Source & flagged code
19 flagged · loading sourcePackage source references child process execution.
claude-bridge/src/claude.jsView on unpkg · L2Source file is highly similar to a previously finalized malicious package; route for source-aware review.
claude-bridge/src/claude.jsView on unpkgPackage source references dynamic require/import behavior.
claude-bridge/src/claude.jsView on unpkg · L2Findings
Affected versions and remediation
This report applies to alvin-bot@5.131.0.
See all versions of alvin-bot for other recorded verdicts.
- Review the evidence and your use of alvin-bot@5.131.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.