ClawHub
Back to skill

Security audit

Stripe

Security checks for vulnerabilities and agentic risk

Overview

This is a clearly disclosed Stripe administration skill with high-impact financial capabilities, but its sensitive behavior is purpose-aligned and guarded by explicit approval instructions.

Install only if you need Stripe administration. Use the CLI/OAuth path, connect the least-privileged Stripe account, verify test versus live mode before any write, require explicit confirmation for charges, refunds, invoice, subscription, or deletion actions, and avoid the raw HTTP fallback unless the CLI cannot be used.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The appendix says direct HTTP writes should send a JSON-encoded body with Content-Type: application/json, but the rest of the skill and Stripe API reference consistently require application/x-www-form-urlencoded for Stripe write operations. This inconsistency can cause failed or malformed write requests in a high-risk financial integration, and may lead operators to improvise unsafe retries or debugging steps around payment, refund, or billing actions.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The direct-HTTP example prints the full response body even though the skill repeatedly warns that Stripe responses may contain regulated payment and personal data and should be minimized. In practice, users often copy examples verbatim, so this encourages unnecessary disclosure of sensitive customer and payment information into terminal logs, transcripts, CI output, or shared debugging artifacts.

Content

No source excerpt is available for this finding.