The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions and does not include any executable scripts, binaries, or configuration files that could run on the host system.
[EXTERNAL_DOWNLOADS]: The instructions recommend verifying legal norms using external sources such as gesetze-im-internet.de, dejure.org, and openJur. These are established and well-known German legal information services used for professional reference. The skill instructs the agent to consult these for verification rather than downloading or executing remote code.
[PROMPT_INJECTION]: The skill contains a 'Quellenregel' (Source Rule) which directs the AI to prioritize official sources over model training data and to avoid hallucinating legal citations. These are safety-enhancing instructions designed to ensure the accuracy of legal advice, not malicious overrides or jailbreak attempts.
[COMMAND_EXECUTION]: No shell commands, subprocess calls, or privilege escalation patterns were detected. The skill focuses solely on text generation and document structuring.
[DATA_EXFILTRATION]: There are no indicators of unauthorized data transmission. The skill mentions handling sensitive legal documents (e.g., contracts, evidence), but only within the context of drafting a document for the user's review. No external endpoints or untrusted domains are targeted for data delivery.

schriftsatzkern-substantiierung