ClawHub
Back to skill

Security audit

Claw Mail

Security checks across malware telemetry and agentic risk

Overview

ClawMail is a coherent email integration, but its setup asks users to run an unverified remote Python script and then stores email credentials locally.

Install only if you trust ClawMail and are comfortable routing agent email through its service. Review the setup script before running it or ask the publisher for a signed or checksum-pinned installer, protect ~/.clawmail/config.json, use a dedicated low-risk inbox, and avoid sending secrets or regulated data unless approved.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The README instructs users to download a Python script over the network and immediately execute it to create credentials, without any integrity verification, pinning, signature check, or warning about the trust boundary. If the remote host, DNS, TLS termination, or distribution path is compromised, users could run arbitrary code locally and expose mailbox credentials or broader system access.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The poll endpoint is documented as returning new messages and marking them as read, but the skill does not prominently warn users about this side effect where they are most likely to act on it. An agent using polling for inspection, debugging, or multi-agent coordination could unintentionally alter mailbox state and cause message loss, missed processing, or broken audit expectations.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill sends email bodies, recipient data, sender metadata, and inbox contents to an external third-party service, but the description does not clearly disclose this privacy and data-handling implication. In agent environments, this can lead to inadvertent transmission of sensitive prompts, personal data, or business communications to infrastructure outside the user's trust boundary.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.