ClawHub
Back to skill

Security audit

Meeting Minutes Task Extractor

Security checks for vulnerabilities and agentic risk

Overview

This skill locally extracts meeting tasks and adds a premium-upgrade payment link, with no evidence of hidden execution, data theft, or persistence.

Install only if you are comfortable with task output including a premium-upgrade link containing the provided user_id. Review any SKILLPAY payment URL environment configuration before use, especially in shared deployments.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
81% confidence
Finding

The skill advertises runnable code but does not declare any explicit tool scope such as allowed tools or permissions, even though static analysis detected environment and file-read capabilities. This creates an implicit trust boundary issue: the implementation may access local files or environment-derived secrets beyond what a user would expect from a meeting-notes task extractor, and the missing declaration prevents policy enforcement and informed review.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill’s stated purpose is meeting-minutes task extraction, but it also generates user-specific payment URLs and returns them in normal output. This creates unnecessary coupling between content processing and monetization, and can expose users to unsolicited billing flows or tracking tied to their user_id even when they only requested note extraction.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Reading billing-related environment variables inside a meeting-notes parser introduces hidden external behavior unrelated to the advertised function. In particular, using a configurable URL template with .format(user_id=user_id) allows deployment-time redirection of users to arbitrary payment endpoints, increasing phishing, tracking, or abuse risk if environment configuration is compromised or unreviewed.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The extractor inserts Chinese defaults such as "待分配", "未设置", and a Chinese fallback task, and it also relies on Chinese-only task cues alongside some English ones. This creates a locale policy issue because the skill forces Chinese-language output regardless of the user's preferred language and does not document opt-in or provide a language selection mechanism.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.