nutmeg-analyse
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data, which represents a potential attack surface for indirect instructions.
- Ingestion points: The skill reads user profile information from
.nutmeg.user.mdand retrieves football documentation and data using themcp__football-docs__search_docstool. - Boundary markers: The instructions do not provide specific delimiters or ignore-instructions wrappers for the ingested data.
- Capability inventory: The skill has access to powerful tools including
Bash,Write,Read, andAgentas defined in its configuration. - Sanitization: There are no explicit sanitization or validation steps described for the data processed during analysis.
- [SAFE]: No evidence of prompt injection, credential harvesting, obfuscation, or unauthorized persistence was found. The skill's behavior and tool access are consistent with its primary purpose of data exploration and visualization.
Audit Metadata