Shop from Samsung - With your creditcard

Security checks across malware telemetry and agentic risk

Overview

This is a real-money CreditClaw payment skill with mostly disclosed behavior, but its Samsung-style marketplace label under-describes broad spending, payment-signing, and payment-link authority.

Install only if you intend to enable a broad CreditClaw payments integration, not just Samsung shopping. Use a dedicated API key, keep auto-approval disabled or very low, set strict merchant/category/domain limits, avoid logging API keys, shipping addresses, and payment headers, and do not enable payment links or x402 signing unless you explicitly need those capabilities.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (19)

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The manifest presents this as a shopping/payment method skill, but the documented API also enables the agent to generate payment links and collect money from arbitrary third parties. That materially expands the trust boundary from 'spend under owner guardrails' to 'operate as a merchant/collector,' which can be abused for fraud, unauthorized billing, or social-engineering workflows not implied by the stated purpose.

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill is described as enabling online shopping, but it also documents x402 agent-to-agent payments and on-chain signing. That is a significant hidden capability increase because signing and transfer workflows can authorize value movement beyond ordinary shopping, exposing users to unintended financial and protocol risks.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
Creating payment links to charge arbitrary third parties is outside the stated shopping use case and gives the agent a monetization capability that can be weaponized for scams, unauthorized invoicing, or collection of funds under false pretenses. Even if operationally legitimate, it is a materially different financial action than purchasing and should not be silently bundled.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The guide instructs agents to transmit a full recipient shipping address, but it does not explicitly warn that this is sensitive personal data being sent to a third-party payment/purchasing service. In an agent skill context, that omission matters because agents may collect and forward address data automatically, increasing privacy, consent, and data-minimization risks if users are not clearly informed.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs periodic authenticated requests that return sensitive financial data across all connected payment rails, including balances, limits, cards, and guardrails, but provides no warning that this transmits wallet metadata to an external service. In an agent-skill context, normalizing silent transmission of financial status increases the risk of overcollection, unintended disclosure, and unsafe autonomous polling with privileged credentials.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
This section directs the agent to issue a state-changing top-up request when balance is low, affecting the user's funds workflow, but omits a clear warning that it triggers an external action on the user's account. In a shopping/payment skill, encouraging autonomous funding requests without explicit consent boundaries can lead to unwanted financial operations, spammy requests, or social-engineering-style pressure on the owner.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation instructs agents to transmit a full shipping address, including recipient name and street address, to the service and downstream merchant without any privacy notice, minimization guidance, or consent warning. Because this skill is specifically designed to place real-world purchases, the data exposure is operationally necessary but still creates a real privacy and compliance risk if agents collect, log, or forward personal data too broadly.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
This documentation instructs an agent to initiate real-money online purchases using a linked payment card, but it does not present an explicit warning that these actions can create real financial charges. In an agent skill context, omission of a prominent charge/consent warning increases the risk of unintended or overly broad spending, especially because the text frames the rail as suitable for 'any online store' and normalizes purchase execution.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
This documentation instructs an agent to request payment signatures and later query wallet balance and transactions, but it does not explicitly warn that these actions can spend real funds and expose sensitive financial/account activity. In an agent-skill context, missing user-facing safety language increases the chance that an agent will autonomously perform monetary actions or reveal wallet metadata without clear consent boundaries.

External Transmission

Medium
Category
Data Exfiltration
Content
## 1. Check Full Status (Recommended)

```bash
curl https://creditclaw.com/api/v1/bot/status \
  -H "Authorization: Bearer $CREDITCLAW_API_KEY"
```
Confidence
89% confidence
Finding
curl https://creditclaw.com/api/v1/bot/status \ -H "Authorization: Bearer $CREDITCLAW_API_KEY" ``` This single endpoint returns your status across **all connected payment rails**, including balance

External Transmission

Medium
Category
Data Exfiltration
Content
Before an expensive self-hosted card purchase, test if it would pass:

```bash
curl -X POST https://creditclaw.com/api/v1/bot/check/rail4/test \
  -H "Authorization: Bearer $CREDITCLAW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "merchant_name": "Amazon", "amount_cents": 5000, "profile_index": 1 }'
Confidence
87% confidence
Finding
curl -X POST https://creditclaw.com/api/v1/bot/check/rail4/test \ -H "Authorization: Bearer $CREDITCLAW_API_KEY" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
### Checkout Request

```bash
curl -X POST https://creditclaw.com/api/v1/bot/merchant/checkout \
  -H "Authorization: Bearer $CREDITCLAW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
89% confidence
Finding
curl -X POST https://creditclaw.com/api/v1/bot/merchant/checkout \ -H "Authorization: Bearer $CREDITCLAW_API_KEY" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
### Request x402 Payment Signature

```bash
curl -X POST https://creditclaw.com/api/v1/stripe-wallet/bot/sign \
  -H "Authorization: Bearer $CREDITCLAW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
88% confidence
Finding
curl -X POST https://creditclaw.com/api/v1/stripe-wallet/bot/sign \ -H "Authorization: Bearer $CREDITCLAW_API_KEY" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
Use the `x_payment_header` value as-is in your retry request:
```bash
curl https://api.example.com/v1/data \
  -H "X-PAYMENT: eyJ0eXAiOi..."
```
Confidence
84% confidence
Finding
https://api.example.com/

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## How It Works

Self-hosted cards use a split-knowledge privacy model. Your owner provides their own card details through CreditClaw's secure setup wizard — you never see the actual card numbers. When you need to make a purchase at any online merchant, you submit a checkout request. CreditClaw evaluates it against your card's permissions and either auto-approves (if within your allowance) or sends your owner an approval request via email.

**Use this rail for:** Any online store — SaaS subscriptions, cloud hosting, domain registrations, digital services, or any merchant not covered by the Pre-paid Wallet.
Confidence
95% confidence
Finding
auto-approve

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
1. You submit a checkout request with merchant and amount details
2. CreditClaw evaluates the request against your card's permissions
3. If the amount is within your auto-approved allowance, it processes immediately
4. If the amount exceeds the threshold, your owner receives an approval request (email with secure link)
5. You poll for the result
6. Once approved, the transaction is recorded
Confidence
97% confidence
Finding
auto-approve

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Allowance Thresholds

Your owner sets a per-profile allowance threshold for each card. Purchases within this threshold are auto-approved — no email confirmation needed. Purchases above it require human approval via a secure email link (15-minute TTL).

Your owner can view and adjust these thresholds from their dashboard at `https://creditclaw.com/app/self-hosted`.
Confidence
94% confidence
Finding
auto-approve

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
**You must follow these rules:**
- If `approval_mode` is `ask_for_everything`, ask your human before any purchase to get their approval. **New accounts default to this mode.** Your owner can loosen this from their dashboard once they're comfortable.
- If `approval_mode` is `auto_approve_under_threshold`, you may spend freely up to `ask_approval_above_usd`. Anything above that requires owner approval.
- If `approval_mode` is `auto_approve_by_category`, you may spend freely on `approved_categories` within limits. All others require approval.
- **Never** spend on `blocked_categories`. These are hard blocks enforced server-side and will be declined.
- Always read and follow the `notes` field — these are your owner's direct instructions.
Confidence
88% confidence
Finding
auto_approve

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
**You must follow these rules:**
- If `approval_mode` is `ask_for_everything`, ask your human before any purchase to get their approval. **New accounts default to this mode.** Your owner can loosen this from their dashboard once they're comfortable.
- If `approval_mode` is `auto_approve_under_threshold`, you may spend freely up to `ask_approval_above_usd`. Anything above that requires owner approval.
- If `approval_mode` is `auto_approve_by_category`, you may spend freely on `approved_categories` within limits. All others require approval.
- **Never** spend on `blocked_categories`. These are hard blocks enforced server-side and will be declined.
- Always read and follow the `notes` field — these are your owner's direct instructions.
- Cache this for up to 30 minutes. Do not fetch before every micro-purchase.
Confidence
88% confidence
Finding
auto_approve

VirusTotal

51/51 vendors flagged this skill as clean.

View on VirusTotal