ClawHub
Back to skill

Security audit

Freeland

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Freeland account integration, but it enables broad implicit access to wallet, card, OTP, purchase, VPN, and invoice actions that users should review carefully before installing.

Install only if you intend your agent to access and operate your Freeland account. Prefer explicit invocation, set clear spending and merchant limits, require confirmation before financial or service changes, and avoid revealing full card credentials, OTPs, inbox messages, or VPN configs unless needed for the active task.

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 (5)

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
freeland_get() {
  local path="$1"
  curl --silent --show-error --get "$BASE_URL" \
    --data-urlencode "path=$path" \
    -H "Authorization: Bearer $FREELAND_API_KEY"
}
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs the agent to retrieve highly sensitive card data and mailbox contents and to hand them to the user during task execution, but it does not require a clear privacy warning, re-authentication step, masking policy, or strict minimization before disclosure. In an agent setting, this increases the risk of overexposing PAN/CVV, OTPs, receipts, and inbox content in chat logs, downstream tools, or to an unauthorized requester if the session context is wrong or compromised.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
freeland_post "/freeman/chat" '{"messages":[{"role":"user","content":"What am I avoiding?"}]}'
```

Mr. Freeman is platform-funded. Do not ask the user for a personal model key for this flow.

### 7. Work with eSIM
Confidence
80% 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.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The default prompt advertises a very broad set of capabilities spanning account access, payment functions, cards, OTP retrieval, chat, telecom purchases, and invoice creation. Because these overlap with common user intents and lack narrow trigger constraints, an agent may invoke the skill in situations where the user did not clearly intend to grant access to sensitive financial or authentication operations, increasing the risk of unintended data exposure or unauthorized actions.

Vague Triggers

Medium
Confidence
97% confidence
Finding
Enabling implicit invocation for a skill with access to wallets, cards, OTP inbox contents, purchases, and crypto invoices is especially dangerous because the agent may auto-select it without a clear, informed user request. In this context, a mistaken or overly broad invocation could expose one-time codes, financial data, or initiate sensitive transactions, making the impact substantially higher than for a low-risk informational skill.

Static analysis

No suspicious patterns detected.