Back to skill

Security audit

PowPow Registrierung — Blase-Konto anlegen, Reisetagebuch auf der Karte verankern, mit digitalen Menschen chatten

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed PowPow account-registration and optional checkout assistant, with sensitive steps scoped to the official service and temporary local state.

Install only if you want an agent to help create a PowPow account. Expect to provide an email, password or generated password, verification code, and optionally an avatar file. For paid upgrades, verify the official PowPow, Alipay, or PayPal destination before completing payment, and run cleanup when finished so the temporary session state is removed.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (25)

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

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
- Kernskript `scripts/powpow-register.mjs`: **null Abhängigkeiten**, benötigt **Node ≥ 18**. Vor dem Start einmal `node scripts/powpow-register.mjs status` ausf

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The README is entirely written as a German-only instruction and workflow description, with no indication that users can choose another language or that the skill is intentionally limited to German-speaking users for a documented reason. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description includes the trigger phrase "hilf mir, ein powpow-Konto anzulegen". Although it contains the product name, the activation wording is still broad help-seeking language and the file does not provide clear exclusion guidance for similar conversational requests beyond a few examples, which could increase unintended invocation risk.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly instructs the agent to invoke OS-level commands (start, open, xdg-open) to launch a locally returned payment page. Even though the stated purpose is limited to opening checkout for the user, this broadens the skill from web/API orchestration into host-command execution, which is a meaningful increase in capability and attack surface if the returned path or invocation context is ever influenced unexpectedly.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill tells the agent to open payment pages via local system commands but does not require a clear user-facing warning or explicit consent immediately before launching a browser/application. That can surprise users, trigger unwanted local side effects, and normalize agent-initiated application launches for financial flows, which is particularly sensitive in a payment context.

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 explicitly sets the skill language to "de", while the rest of the listing includes Chinese text and gives no indication that German is optional or required for a region-specific purpose. This creates a natural-language locale constraint without user opt-in or justification, which matches the language/locale policy violation category.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

These lines explicitly require maintaining the existing tone and workflow language, while the file itself is fully German-language. There is no indication that the user can opt into another language or choose locale, which can violate language/locale policy for a general-purpose skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

This file imposes a specific language/locale presentation on users and operators through its natural-language instructions and quoted interface text. Under the policy, forcing a language without opt-in or documented justification is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Line L03 says to use the screenshots "Bei Bedarf" and whenever users do not know Bubble or want to see the result, which leaves activation scope open-ended and subjective. The file does not define clear boundaries, explicit trigger phrases, or negative examples to distinguish when the skill should send screenshots versus when it should not.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

The quick-access section hardcodes German user-facing phrases and interaction examples, and the document provides no indication that the language should adapt to the user's preference. This can violate a language/locale policy if the skill is expected to respect user-selected language rather than forcing German by default.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documentation explicitly limits what is saved in the temporary state file. However, later saveState calls persist several extra fields beyond that list, so the comment materially understates what local state is retained.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The top-level description frames the experience as a simple in-chat registration flow where the user only supplies basic details and retrieves a code once. However, the same file later documents materially broader behavior around creating orders, opening external payment checkout, and tracking payment completion, which contradicts the simplified claim about what the user interaction entails.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

Line L004 states the script has no dependencies, while L080 specifies using platform-specific external commands (start, open, xdg-open) as part of the execution flow. This is a documentation-level contradiction because the file presents the script as dependency-free while relying on external system facilities outside Node itself.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

This markdown file is primarily written in German, but line L28 presents a mandatory user-facing message in Chinese. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is explicitly justified, which this file does not do.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

Lines L71-L84 enumerate many expected platform errors exclusively via Chinese message text, while the rest of the file is German. This suggests a fixed-language experience without explicit opt-in or documentation that the skill is limited to a Chinese-localized platform, which falls under the locale-policy check.

Content

No source excerpt is available for this finding.