Back to skill

Security audit

PowPow रजिस्ट्रेशन — बुलबुला रजिस्टर करीं, यात्रा-बखान के नक्शा पर पिन करीं, डिजिटल मनई से बातचीत करीं

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for PowPow account registration, but it handles login tokens and payment/order data with under-disclosed local state and automatic checkout-opening behavior that merits review.

Install only if you intend an agent to help create a PowPow account and optionally initiate subscription checkout. Treat generated login links as active session links, run cleanup when finished, and review any checkout file or browser-opening step before proceeding with payment. Avoid sharing terminal output from status if it contains order, subscription, or approval URL details.

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
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (24)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 65)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 76)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 84)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 111)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 122)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 126)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 133)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 134)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 139)May include surrounding context.

md
- मुख्य स्क्रिप्ट `scripts/powpow-register.mjs`: **कौनो डिपेंडेंसी नईखे**, **Node >= 18** चाहीं। शुरू करे से पहिले एक बेर `node scripts/powpow-register.mjs stat

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill description is written entirely in Bhojpuri/Hindi and presents the agent behavior and user flow in that language, but does not state that the user can choose another language or that the skill is limited to a specific locale for compliance or regional reasons. This can violate a language/locale policy when a skill effectively forces one language without user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs the agent to execute a system-level open command on a locally generated file, which crosses from advisory behavior into host action without an explicit confirmation gate immediately before execution. Even if the file is expected to be a payment page produced by the companion script, this creates a risky pattern: a compromised script, path manipulation, or unexpected file content could cause the agent to launch untrusted local content or trigger unintended browser/system behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The search keywords are broad and generic enough to attract invocations outside a narrowly scoped registration/travel-social use case, and there are no trigger constraints or exclusions to reduce accidental matching. In a skill that advertises agent-driven registration, code sending, and ordering, overly broad discovery terms increase the chance of unintended activation and user confusion that could expose users to unsafe automation flows.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest hard-codes the language as Bhojpuri without indicating user choice, fallback behavior, or justification, which can misroute users into a locale they do not understand. For a workflow involving registration, one-time codes, login, and optional payment, language mismatch can cause misunderstanding of sensitive steps and increase the risk of user error or consent issues.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The entire skill file is written in Hindi/Bhojpuri-script instructions and does not indicate that the user can choose another language or locale. The policy scope includes language or locale constraints, and this documentation appears to force a specific language presentation without opt-in or justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The file is written entirely in Bhojpuri/Hindi-script content and does not indicate that users can choose another language or that this language restriction is required for a region-specific purpose. Under the policy, forcing a specific language without opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This markdown file presents all operational instructions and troubleshooting in Hindi/Bhojpuri-script content and does not indicate that other languages are available or that the language choice is user-selected. Under the stated policy, forcing a specific language without opt-in is a natural-language policy violation unless the locale constraint is justified or optional.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The guidance tells the agent to send similar screenshots directly whenever the user lacks a clear concept or wants to see the product, which is broad and weakly scoped. In practice this can cause unnecessary or excessive file sharing, increase the chance of sending irrelevant internal assets, and bypass a more explicit user-consent or relevance check.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation says the temp state file only saves email, username, nickname, JWT token, order number, and workflow step. In practice, later saveState calls also persist fields such as codeSentAt, registeredAt, userId, provider, package, amount, subscriptionId, approveUrl, plan details, and timestamps, which directly contradicts the stated storage scope.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The status command claims not to expose credentials, but it returns the full persisted state with only token removed, which can include sensitive URLs such as approveUrl. In this skill, login-link generation also embeds the live JWT in a URL query parameter, so exposing stored URLs or similar state via status can leak usable authentication or payment/session artifacts to logs, operators, or other local observers.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The file mandates a fixed conversational voice and style for all interactions, indicating the agent should consistently use a particular service tone. This is a language/locale-style constraint presented as a requirement, with no user opt-in or alternative offered.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The description includes interface text such as 「开始对话」 inline, but the document does not state any user language preference, opt-in, or locale justification. This may create a language/locale policy issue because the skill presents language-specific content without documenting choice or region-specific scope.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The HTML generated for the PayPal redirect page hard-codes lang="en" and English-only title text. This is a natural-language locale choice embedded in the skill without any user opt-in or documented justification, which can violate language/locale policy requirements.

Content

No source excerpt is available for this finding.