The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data from spreadsheets using the office-mcp tools.
Ingestion points: Data is ingested through the read_xlsx and analyze_spreadsheet tools referenced in the SKILL.md metadata.
Boundary markers: The skill lacks explicit instructions or delimiters to isolate processed spreadsheet data from the agent's core instructions, potentially allowing embedded text to influence agent behavior.
Capability inventory: The skill has the ability to write data via the create_xlsx tool. It does not have network access or shell execution capabilities.
Sanitization: No specific sanitization or validation logic is defined for the content extracted from spreadsheets.

Lead Qualification