The Agent Skills Directory

[SAFE]: The skill consists of markdown-based instructions for legal analysis and drafting. It does not include any scripts, executable code, or binary files.
[SAFE]: No evidence of data exfiltration or unauthorized network access was found. The mentioned legal databases (e.g., gesetze-im-internet.de, openJur) are standard, well-known resources for German legal research and are used here for reference rather than automated scraping.
[SAFE]: The skill contains no prompt injection techniques, obfuscated content, or persistence mechanisms. All instructions are transparent and aligned with the stated purpose of legal contract review.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external legal documents and evidence provided by users.
Ingestion points: Processes user-supplied documents such as contracts, emails, and expert reports (SKILL.md).
Boundary markers: No specific boundary markers or sanitization instructions are defined for the input data.
Capability inventory: The skill is limited to generating text-based outputs, such as legal memos and internal directives. It does not call external tools for file writing or network operations.
Sanitization: No specific sanitization or escaping of external content is specified. The risk is considered negligible as the agent acts purely in an advisory capacity within a text-only context.

rechtsabteilung-ki-saas-vertrag-werkvertrag