dsgvo-audit

• Outcome: Review
• Latest audit: Oct 12, 2026 · 6:38 AM UTC
• Version: 0.1.0

Overview

The skill is a coherent GDPR audit workflow, but it can broadly scan sensitive repositories, write detailed reports to the user's Desktop, and spawn subagents after a broad trigger like "compliance check."

Install only if you intentionally want a German/EU GDPR audit of the current repository. Confirm the audit scope first, review where reports will be written, and avoid running it on repositories containing secrets, customer data, or confidential code unless Desktop report storage and any subagent review are acceptable.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Rogue Agent: Self-Modification, Session Persistence
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (10)

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · README.md (reported line 29)May include surrounding context.

md

### Phase 0: PII Inventory (Art. 4, 30 GDPR)
- **Database Schema Audit**: Scans database definitions (e.g., Prisma schemas) to build a registry of Personal Identifiable Information (PII) including names, emails, phone numbers, addresses, IP addresses, and device identifiers.
- **Sensitive Data Detection**: Identifies special categories of data (Art. 9 GDPR) and sensitive access credentials (such as RustDesk IDs and passwords).
- **Data Flow Mapping**: Traces the lifecycle of PII from system entry points, through internal APIs, to third-party integrations.
- **Log Leakage Audit**: Scans Fastify logs, error outputs, and Server-Sent Events (SSE) payloads for unintended PII logging.

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 409)May include surrounding context.

md

### Phase 0: PII Inventory (Art. 4, 30 GDPR)
- **Database Schema Audit**: Scans database definitions (e.g., Prisma schemas) to build a registry of Personal Identifiable Information (PII) including names, emails, phone numbers, addresses, IP addresses, and device identifiers.
- **Sensitive Data Detection**: Identifies special categories of data (Art. 9 GDPR) and sensitive access credentials (such as RustDesk IDs and passwords).
- **Data Flow Mapping**: Traces the lifecycle of PII from system entry points, through internal APIs, to third-party integrations.
- **Log Leakage Audit**: Scans Fastify logs, error outputs, and Server-Sent Events (SSE) payloads for unintended PII logging.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

The trigger phrase compliance check is very broad and likely to match unrelated user requests, causing accidental activation of a heavy audit workflow. In this skill, accidental activation is more concerning because invocation leads to extensive repo scanning, command execution, and file writes, amplifying the effect of an imprecise trigger.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The trigger list again includes compliance check without constraints, reinforcing a broad auto-activation surface. Because this skill performs multi-phase scanning and output generation, an overbroad trigger can cause unintended inspection of sensitive repositories when the user meant a different kind of review.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill directs writing multiple reports to the user's Desktop outside the repository, creating a filesystem write side effect not clearly justified by the manifest. Writing audit outputs to a global Desktop path can leak repository-sensitive findings to an unintended location, overwrite user expectations about storage, and bypass tighter workspace boundaries.

Content

No source excerpt is available for this finding.

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 80% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 47)May include surrounding context.

md

### Pre-flight (before any phase)

1. **Task list**: Create a TaskCreate entry for each phase. Mark each `in_progress` when starting, `completed` when the phase file is verified. Never mark a phase complete without file verification.

2. **Resolve output path**:

Context-Inappropriate Capability

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

The skill explicitly instructs broad shell command execution (bash/PowerShell) and repository-wide ripgrep scans across the target codebase. Even though the stated purpose is a compliance audit, this materially expands the skill from descriptive guidance into active command execution over user data and source, which increases risk of over-collection, accidental exposure of secrets/PII in outputs, and misuse in environments where command execution is sensitive.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 88% confidence

Finding

The title and throughout the task list mandate a DSGVO/GDPR audit framed in German legal terminology, including German-specific outputs and labels, without offering a language or locale choice. This can violate language/locale policy when the skill is used in broader contexts and no explicit opt-in or region restriction is stated in the file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 91% confidence

Finding

The instructions require checks for 'Impressum completeness (DDG §5)' and 'TTDSG §25', which are Germany-specific legal requirements, but the file does not explicitly state that the skill is restricted to German websites or German legal review. Requiring this locale-specific framework without opt-in or clear justification is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low

• Category: Not specified by scanner
• Confidence: 82% confidence

Finding

The Challenge Pass instructs spawning fresh subagents to validate findings, adding an orchestration/data-sharing behavior beyond the core stated skill purpose. This can increase data propagation by sending finding text and repository paths to additional agents, raising confidentiality and control concerns even if used for quality assurance.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.