ClawHub
Back to skill

Security audit

Health Insurance

Security checks for vulnerabilities and agentic risk

Overview

This instruction-only health insurance skill is purpose-aligned and local, but users should be careful about what sensitive insurance evidence they choose to save.

Before installing, decide whether you want a local health-insurance folder. Only approve memory or artifact storage for information you want retained, and redact policy numbers, government IDs, payment details, and unnecessary personal information from screenshots or confirmations.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
### 6. Execute Enrollment With Evidence
Use `enrollment-playbook.md` to define exact actions, deadlines, and proof artifacts.
Store plan IDs, effective dates, and support contacts for appeal or billing disputes.
Never claim enrollment is complete without confirmation evidence.

### 7. Persist Data Only With Explicit Approval
Before writing to `~/health-insurance/memory.md`, ask for explicit confirmation.
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
95% confidence
Finding
The instruction to save screenshots or confirmation artifacts can lead to storage of documents containing sensitive personal, policy, or insurance identifiers without any guidance on minimization, secure storage, or redaction. In a health-insurance enrollment context, these artifacts may include regulated personal and health-related information, increasing the risk of privacy exposure if stored insecurely or shared broadly.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Content
```bash
mkdir -p ~/health-insurance/{comparisons,renewals,notes}
touch ~/health-insurance/memory.md
chmod 700 ~/health-insurance
chmod 600 ~/health-insurance/memory.md
```
2. If approved and `memory.md` is empty, initialize from `memory-template.md`.
Confidence
80% confidence
Finding
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Content
mkdir -p ~/health-insurance/{comparisons,renewals,notes}
touch ~/health-insurance/memory.md
chmod 700 ~/health-insurance
chmod 600 ~/health-insurance/memory.md
```
2. If approved and `memory.md` is empty, initialize from `memory-template.md`.
3. Continue with the user request immediately after setup.
Confidence
80% confidence
Finding
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Static analysis

No suspicious patterns detected.