Security audit
Security checks for vulnerabilities and agentic risk
The skill is mostly coherent for PowPow account setup, but it handles account credentials, session tokens, and payment flows with under-disclosed local state and risky browser-opening behavior.
Review this carefully before installing. Use it only if you are comfortable letting the agent help create a PowPow account, send verification requests, hold a temporary session token locally, and create payment checkout sessions. Do not provide payment credentials to the agent, confirm before any order is created or browser window is opened, and run cleanup when finished.
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
Referenced artifact was not completely inspected
- Iwe pataki `scripts/powpow-register.mjs`: **laisi awọn igbẹkẹle**, o nilo **Node >= 18**. Ṣaaju ki o to bẹrẹ, ṣiṣẹ lẹẹkan `node scripts/powpow-register.mjs st
The README describes an agent performing sensitive account actions—registration, profile image upload, payment-order creation/tracking, and one-click login—without a prominent, explicit warning about privacy implications, account changes, and when user confirmation is required. In this context, users may unknowingly authorize disclosure of personal data or account-affecting operations, increasing the risk of unintended registration, token exposure, or payment-related mistakes.
A skill document that forces a specific language can violate language/locale policy when users are not given a choice. This README provides all operational and safety information in Yoruba only, with no indication of alternative language support or user opt-in.
The skill instructs the agent to handle sensitive registration data, email verification, login-link generation, and payment initiation, and it explicitly relies on environment variables and outbound requests to the PowPow service. Without an explicit tool/permission scope, the agent may be granted broader-than-necessary access at runtime, increasing the chance of unintended network use or exposure of secrets during account creation and payment flows.
The trigger list includes broad natural-language phrases for signup help and post-registration issues, which can cause the skill to activate in loosely related conversations. Because this skill collects credentials, email addresses, verification codes, and may initiate payment flows, accidental invocation creates a meaningful risk of prompting users for sensitive data in the wrong context.
The manifest description and the full user-facing instructions are presented in Yoruba, and the file does not indicate that users may choose another language or opt in to this locale. A fixed-language requirement without choice can violate language/locale policy when the skill is intended for general users rather than a clearly documented region-specific audience.
The listing metadata markets broader social and map-based capabilities than the described registration/login-only function, creating a scope mismatch. This can mislead users and reviewers about what the skill does, weakening informed consent and making it easier for account-handling automation to be invoked under a broader, less precise description.
The notes describe agent-driven automation for sending verification codes, registering accounts, and placing orders, but do not define strict activation boundaries or narrow trigger conditions. For a skill that handles identity, email verification, uploads, and optional payments, ambiguous scope increases the risk of overreach, unintended actions, or social-engineering-friendly misuse.
The entire skill reference is written exclusively in Yoruba and does not indicate that users may choose another language or locale. This creates a natural-language policy concern because it imposes a specific language experience without documented opt-in or justification as a region-specific tool.
The skill instructs the agent to write server-returned paymentHtml to a temporary local HTML file and open it via a system command. That combines untrusted remote content with local file creation and OS-level execution behavior, creating risk of unsafe browser launch flows, unexpected local persistence, and abuse if the returned HTML is ever malicious or compromised. In a payment context this is more dangerous because users may be primed to trust the opened page and enter credentials or approve transactions.
The PayPal flow similarly directs the agent to create a local HTML file and open it in a browser without explicit user warning or consent. Even if the intent is convenience, silently creating files and launching browser content from workflow data increases the attack surface for phishing, unsafe redirects, and user confusion about whether the page is local, agent-generated, or truly PayPal-hosted. Because this is an authentication/payment flow, the trust sensitivity is elevated.
The entire skill file is written in Yoruba, while embedded examples include Chinese strings, and there is no natural-language note offering an alternate language or stating that the skill is intentionally region- or locale-specific. This can violate a language/locale policy when users or operators are expected to consume the instructions without prior opt-in to that language.
The guidance says to use screenshots whenever they may help during execution and includes broad conditions like when a user does not understand what Bubble is. That can cause the skill to activate or inject product-marketing content into ordinary conversations without a tight user-intent boundary, increasing the chance of unintended invocation and irrelevant content delivery.
The header comment materially understates what is written to the state file, while the code persists additional personal and payment-related data such as phone, provider, amount, subscription identifiers, approval URLs, timestamps, and account status. In a registration/payment automation skill, misleading data-retention claims are security-relevant because operators may handle the tool under incorrect assumptions, increasing privacy, compliance, and local data-exposure risk if the temp file is accessed by another process or retained longer than expected.
The manifest sets the skill language to "yo" while other visible text in the file mixes Yoruba and Chinese, and there is no indication that users can opt into another language or that the locale restriction is required for a region-specific purpose. This can violate language or locale policy when a skill enforces one language without user choice or justification.