ClawHub
Back to skill

Security audit

Sales Memory

Security checks for vulnerabilities and agentic risk

Overview

This skill openly provides sales memory through BlueColumn, but it can automatically send and persist customer sales details to a third-party service without enough scoping or consent guidance.

Review before installing. Use only with an approved BlueColumn account and clear team policy for what sales data may be stored. Avoid secrets, regulated data, unnecessary personal details, and confidential deal information unless your organization has approved BlueColumn retention, access, and deletion controls.

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)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill is designed to send and persist sales conversation history to an external service, but it provides no notice, consent flow, retention guidance, or data-sensitivity restrictions. Because the examples include identifiable customer context and deal history, this can expose personal data, confidential sales information, and regulated business records to a third-party system without adequate transparency.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The activation guidance is overly broad for a skill that can recall and persist remote sales memory, making it more likely to trigger on ordinary sales conversations without explicit user intent. In this context, over-invocation increases the chance of unnecessary access to prior customer context and unintended storage of sensitive business or personal information.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

The example explicitly posts sales notes to an external endpoint, which is a real data egress path rather than a purely local operation. While external transmission is expected for this skill's functionality, it is still security-relevant because the transmitted content may contain customer identifiers, objections, follow-up dates, and other sensitive commercial context.

Content

Scanner excerpt · SKILL.md (reported line 15)May include surrounding context.

Store

bash
curl -X POST .../agent-remember \
  -H "Authorization: Bearer <key>" \
  -H "Content-Type: application/json" \
  -d '{"text": "Jane's objection: price. Addressed with annual discount. Follow-up set for Aug 5.", "title": "sales-memory - note"}'