abo-kuendigung-fitness-streaming
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a workflow for processing external, untrusted documents such as contracts and Terms and Conditions.
- Ingestion points: Section 1 (SKILL.md) directs the agent to read user-provided documents.
- Boundary markers: The instructions do not define delimiters to separate document content from the prompt context.
- Capability inventory: The skill does not include any tools, scripts, or active capabilities like network access or shell execution.
- Sanitization: No sanitization or validation of the input data is described.
- [NO_CODE]: The skill is composed entirely of markdown instructions and contains no code snippets, shell commands, or external package dependencies.
Audit Metadata