The Agent Skills Directory

[SAFE]: The skill defines a structured legal analysis workflow for patient rights under § 630g BGB and Art. 15 GDPR, utilizing a sub-agent architecture (researcher, drafter, reviewer).
[EXTERNAL_DOWNLOADS]: The skill references official legal documentation from 'gesetze-im-internet.de', 'eur-lex.europa.eu', and 'curia.europa.eu'. These are well-known official resources for German and European law and are documented neutrally.
[DATA_EXPOSURE]: No hardcoded credentials, secrets, or sensitive local file paths were found. The skill instructions focus on processing anonymized or user-provided legal case data.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted fact patterns to generate legal documents.
Ingestion points: Fact patterns and case details are processed through the 'Eingaben' section in SKILL.md and the 'fact_pattern' field in test.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt templates.
Capability inventory: The skill uses a drafter agent to generate claim letters and legal statements based on the provided input.
Sanitization: No validation or sanitization of input data is specified in the instructions.

patientenrechte-akteneinsicht