ClawHub
Back to skill

Security audit

Client Brief Routing Desk and Finance Review Mailer

Security checks for vulnerabilities and agentic risk

Overview

The skill is a small, disclosed routing and finance-review messaging helper, with the main caveat that users should confirm before any outbound message is sent.

Install only if you want an agent to help choose routing contacts and send finance-review messages. Configure or instruct the agent to ask for confirmation before calling the send tool, especially when recipient addresses or account-review details are involved.

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 (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The skill is described with a broad activation scope for 'routine client communications work' while bundling both contact selection and outbound messaging. This increases the chance the agent invokes a message-sending capability in situations where the user did not clearly intend an external communication, creating a pathway for unintended disclosure or unauthorized outbound actions.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The documentation states the skill can send finance-review messages but does not prominently warn users that invoking the skill may cause an outbound communication to a recipient. In an agent context, missing disclosure around external side effects is dangerous because users may believe they are only selecting a contact, while the skill can escalate to sending a message that could expose internal account or finance-review information.

Static analysis

No suspicious patterns detected.