excel-dcf-modeler
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input and interpolates it into an Excel workbook using the Excel MCP server tool. * Ingestion points: User interaction during the input gathering phase described in SKILL.md. * Boundary markers: The instructions do not provide delimiters or safety warnings for the agent to sanitize the data before passing it to the tool. * Capability inventory: The skill utilizes the Excel MCP server to create workbooks and write to the local file system. * Sanitization: No evidence of validation or filtering for Excel formula triggers (e.g.,
=,+,-,@) is present in the instructions. - [NO_CODE]: The skill consists entirely of instructional markdown and reference files and does not contain any executable scripts, binaries, or automated installation commands.
Audit Metadata