agg-entschaedigung
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user facts that are subsequently processed by agent workflows without explicit boundary protection.\n
- Ingestion points: Case description inputs in
SKILL.mdandtest.md.\n - Boundary markers: Absent; there are no specific instructions provided to the agent to treat external data as untrusted or to use delimiters.\n
- Capability inventory: Triggering sub-agent tasks and executing a local calculation script (
scripts.legal_calc.cli).\n - Sanitization: Absent; user input is interpolated into the workflow without evidence of filtering.\n- [COMMAND_EXECUTION]: The skill invokes a local Python script for deterministic legal date calculations.\n
- Evidence: Shell commands using
python -m scripts.legal_calc.cliin the body ofSKILL.md.\n - Context: Used to calculate two-month and three-month legal periods as defined by German law.
Audit Metadata