The Agent Skills Directory

[SAFE]: The skill's primary function is to guide an agent through a legal analysis workflow based on German civil law (BGB). The instructions are consistent with professional legal assistance and do not contain attempts to bypass safety filters or perform unauthorized actions.
[SAFE]: The skill mentions external legal databases (gesetze-im-internet.de, dejure.org, openJur, and official court databases) for verifying statutory norms and case law. These are recognized, authoritative sources in the legal domain and are used for verification purposes rather than downloading executable content.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external documents such as contracts, administrative acts, and expert reports provided during a legal case. While this creates an attack surface where instructions could be embedded in ingested files, the skill's narrow focus on German legal structures and predefined templates significantly limits the risk of the agent deviating from its intended purpose.
Ingestion points: Processes user-provided documents like administrative acts, contracts, and case files (referenced in SKILL.md).
Boundary markers: No explicit technical delimiters for external data are defined, but the instructions emphasize strict adherence to legal norms and verified sources.
Capability inventory: The skill generates text-based protocols, legal notices, and memos. It does not possess capabilities for file writing, system command execution, or non-whitelisted network operations.
Sanitization: No explicit sanitization or filtering of external content is mentioned.

software-mangel