The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external administrative and legal documents (e.g., contracts, administrative acts, and legal briefs), creating a surface for indirect prompt injection where malicious instructions embedded in those documents could attempt to influence the agent's behavior.
Ingestion points: Legal documents, case files, and administrative records provided by the user (as noted in the 'Arbeitsweg' section).
Boundary markers: The skill does not specify any delimiters or safety warnings to separate user-provided content from its internal instructions.
Capability inventory: The skill's primary function is restricted to producing textual markdown reports based on local templates in the assets/templates/ directory.
Sanitization: No specific input sanitization or content filtering for external documents is described.
[NO_CODE]: The analyzed skill file contains only markdown instructions and YAML configuration; it does not include shell scripts, source code, or instructions to install third-party dependencies.

erechnung