nutmeg-compute
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from external sources such as football stat APIs, web pages, and downloaded files, which creates an inherent attack surface for indirect prompt injection.
- Ingestion points: The skill instructions in
SKILL.mdexplicitly reference processing content from external API responses, web scraping (e.g., Understat), and documentation retrieved via themcp__football-docs__search_docstool. - Boundary markers: The skill includes a 'Security' section that provides explicit instructions to the agent to treat all external content as untrusted and to avoid using it to modify system prompts or tool configurations.
- Capability inventory: The skill utilizes tools with broad capabilities, including
Bash,Write, andRead. - Sanitization: The instructions include a requirement for the agent to validate data shapes and schemas before processing external content.
Audit Metadata