ClawHub
Back to skill

Security audit

Willhaben

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for creating Willhaben listings, but it can use a logged-in browser and publish public ads after confirmation.

Install only if you want the agent to help operate your Willhaben account through a logged-in browser. Review each draft carefully before approving publication, and avoid storing precise location details in the preferences file unless you are comfortable keeping them locally in the skill folder.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest description includes broad triggering language such as "sell something" and "mentions Willhaben," which can cause the skill to activate on vague marketplace-related conversations rather than an explicit request to post or manage a listing. Because this skill can use saved credentials and browser automation to create public listings, over-invocation raises the risk of unintended account actions, accidental publication, and premature collection or use of user photos and listing data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill operates using a saved browser profile with an already authenticated Willhaben session, but the skill description does not clearly warn the user that account credentials and browser automation will be used to create or modify public listings. This weakens informed consent and makes the skill more dangerous in context, because a user may invoke it without realizing it can act directly on their marketplace account and publish content publicly.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This manifest-style JSON sets "language": "de" as a fixed default, which can constitute a language/locale policy violation when no alternative or opt-in mechanism is described. Because the file contains no indication that language selection is optional or region-justified, it risks enforcing a specific locale on users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

Several setup prompts are hardcoded in German, such as "Wo bist du?" and "Preisstrategie?", with no indication that the user can choose another language. This creates a language/locale policy issue because the skill imposes a specific language without opt-in or documented justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The setup persists personal data including the user's district/city and preferences to a local JSON file without any explicit notice, consent flow, retention policy, or guidance on protecting that file. In the context of a marketplace-listing skill, this data is directly tied to a user's physical area and selling behavior, so unauthorized local access or accidental disclosure could expose sensitive personal information.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.