ClawHub
Back to skill

Security audit

Personal CRM

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local personal CRM that clearly discloses storing relationship notes and optional calendar/email use, with no evidence of hidden execution or external sharing.

Before installing, be aware that this skill is meant to create a durable local record of personal contacts, birthdays, emails, phone numbers, and conversation history. Use it with explicit intent, avoid importing third-party data you do not want retained, and review any calendar, email, Facebook export, or future transcript-derived entries before saving them.

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)

Context Leakage

High
Category
Data Exfiltration
Confidence
76% confidence
Finding

Auto-extracting conversation details from meeting transcripts increases the risk of ingesting sensitive third-party information into the CRM without adequate minimization or consent. In the context of a relationship manager that stores long-lived personal notes, this can leak context across tools and preserve details that users or contacts did not intend to be retained in this repository.

Content

Scanner excerpt · SKILL.md (reported line 306)May include surrounding context.

md
| **Calendar** | Pre-meeting context, post-meeting prompts, birthday import |
| **Email** | Forward emails to log touchpoints, auto-extract email addresses |
| **Gift Tracker** (separate skill) | Link gift history to contacts, suggest gift ideas |
| **Meeting Transcripts** (future) | Auto-extract conversation details from tools like Granola |

---

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases for logging touchpoints are very broad, natural utterances that a user might say in ordinary conversation without intending to invoke this skill. That can cause unintended activation, prompting the agent to capture or store sensitive relationship details when the user was only chatting casually, which is a privacy and consent issue in a CRM-style skill handling personal data.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.