microsoft-excel
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits data from Excel workbooks to the Maton infrastructure (gateway.maton.ai) as part of its core functionality. While this is the intended purpose, it involves sending potentially sensitive spreadsheet content to a non-whitelisted external domain.
- [COMMAND_EXECUTION]: The documentation includes multiple Python code blocks intended for execution to interact with the API. These snippets use urllib.request and the requests library to perform network operations and process environment variables.
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves data from external Excel workbooks stored in OneDrive or SharePoint, which represents a potential ingestion point for untrusted content that could influence agent behavior.
- Ingestion points: API operations documented in SKILL.md for reading cell ranges, used ranges, and table rows.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the tool usage guidelines.
- Capability inventory: The skill provides broad capabilities to read, write, update, and delete worksheets, tables, and cell values within the linked Microsoft account.
- Sanitization: There is no evidence of data sanitization or validation performed on the retrieved Excel content before it is processed by the agent.
Audit Metadata