ClawHub
Back to skill

Security audit

Sop Generator

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only SOP-writing skill with some privacy considerations around optional third-party automation, but no hidden execution or persistence.

Installers should treat this as a benign SOP documentation helper. Before using its automation ideas, review transcripts for secrets, customer data, regulated information, or private internal details, and use only approved tools with appropriate sharing and retention settings.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list includes broad natural-language phrases like "document this process" and "make this team-ready," which can cause the skill to activate in contexts broader than intended. Overbroad activation increases the chance of unsolicited processing of user content, including sensitive operational data, especially when paired with downstream automation features.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Step 1: Understand What Makes a Good SOP

A good SOP is:
- **Actionable:** Anyone following it can complete the task without asking questions
- **Complete:** Covers every step, even the obvious ones
- **Structured:** Same format every time so it's easy to scan
- **Maintained:** Updated when the process changes
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Step 1: Understand What Makes a Good SOP

A good SOP is:
- **Actionable:** Anyone following it can complete the task without asking questions
- **Complete:** Covers every step, even the obvious ones
- **Structured:** Same format every time so it's easy to scan
- **Maintained:** Updated when the process changes
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The automation section recommends sending transcripts to external services and APIs (e.g., Loom integrations, Claude API, Notion, Slack) but provides no privacy, consent, retention, or data-classification warning. Transcripts can contain credentials, customer data, internal procedures, or regulated information, so forwarding them to third parties without safeguards can lead to data exposure or compliance violations.

Static analysis

No suspicious patterns detected.