dsv-meldung-lfdi-bw
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially untrusted external documents to draft legal reports, which creates a surface for indirect prompt injection.
- Ingestion points: The skill instructions (in SKILL.md) direct the agent to collect and examine external documents such as administrative acts, contracts, briefs, protocols, and expert reports.
- Boundary markers: There are no specific instructions or delimiters defined to separate user-provided document content from the agent's core instructions, nor are there warnings for the agent to ignore instructions embedded within those documents.
- Capability inventory: The skill's capabilities appear limited to text generation and structured reporting. There are no identified tools for network transmission, shell command execution, or sensitive file writing that could be abused by an injection.
- Sanitization: The instructions do not prescribe any methods for sanitizing, filtering, or escaping content found within the ingested legal documents before they are processed by the LLM.
Audit Metadata