grundschuld-sicherungsrecht
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by ingesting untrusted data from the user to drive agent actions and tool arguments. * Ingestion points: User-provided values for property details, loan amounts, and contract wording (Sicherungszweckerklärung) in SKILL.md. * Boundary markers: The skill does not define specific delimiters or instructions to treat external data as untrusted. * Capability inventory: The skill utilizes sub-agents and a local Python calculation tool. * Sanitization: No sanitization or validation of input data is defined.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python module (scripts.legal_calc.cli) via the command line to perform financial calculations. * Evidence: The 'Deterministische Berechnung' section in SKILL.md provides bash command templates with user-derived arguments.
- [EXTERNAL_DOWNLOADS]: The skill references and fetches information from reputable official legal websites. * Evidence: The Researcher agent fetches statutes and rulings from gesetze-im-internet.de and dejure.org, which are established official sources for German law.
Audit Metadata