The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data provided by the user, such as legal contracts, administrative acts, and expert reports, for the purpose of analysis.
Ingestion points: The workflow involves gathering and reviewing documents like "Verwaltungsakte", "Vertragsurkunden", and "Schriftsätze" (found in SKILL.md).
Boundary markers: No specific delimiters or instructions to ignore embedded content are present in the guidelines.
Capability inventory: The skill is limited to textual analysis and report generation; it does not possess capabilities for arbitrary command execution, file system modification, or network exfiltration.
Sanitization: There are no explicit instructions to sanitize or validate the content of the ingested legal documents.
[NO_CODE]: The skill contains no executable scripts (Python, Node.js, Shell, etc.) or binary components, which significantly limits the potential for technical exploitation.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and official German legal databases for information verification.
Trusted Sources: It suggests cross-referencing information using official domains and services such as gesetze-im-internet.de, dejure.org, and openJur. These references are used for legitimate legal research purposes.

arbeitnehmer-software-beta-pilot