ClawHub
Back to skill

Security audit

DoctorClaw Email Digest

Security checks for vulnerabilities and agentic risk

Overview

This email-digest skill is coherent and purpose-aligned, but users should be careful because it can read inbox contents and optionally deliver summaries through external channels.

Install only if you are comfortable giving the agent read access to unread emails. Use a narrow max-email limit, avoid external delivery channels for sensitive inboxes unless needed, and enable send access or scheduled cron runs only after confirming the approval and delivery behavior you want.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill advertises on-demand use whenever an inbox 'feels overwhelming,' which is a vague, natural-language trigger surface that can overlap with ordinary conversation. In an agent environment, broad trigger wording can cause unintended invocation of a high-privilege workflow that reads inbox contents and drafts replies, exposing sensitive email data or causing unexpected side effects.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly supports delivering digests via Telegram/Discord or file output but does not clearly warn that email subjects, previews, action items, and draft replies may be transmitted to third-party channels. Given that inbox contents often contain sensitive business and personal information, this omission can lead users to unknowingly exfiltrate confidential data outside the email system.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The example trigger 'Check my email' is extremely generic and likely to appear in normal user conversation, making accidental activation plausible. Because this skill accesses unread messages and may prepare outbound replies, unintended invocation can leak private message content into the chat or downstream delivery channels.

Static analysis

No suspicious patterns detected.