The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external project data from various formats including CSV, Excel, and JSON. This creates a potential surface where malicious instructions could be embedded within the data fields to influence the agent's behavior during reporting or analysis.
Ingestion points: Data provided by the user or from local files in CSV, Excel, or JSON formats (as specified in instructions.md).
Boundary markers: None explicitly defined in the provided Python logic to delimit untrusted data from instructions.
Capability inventory: The skill has filesystem write access to export Excel spreadsheets and HTML dashboards (SKILL.md).
Sanitization: The skill relies on standard pandas and openpyxl parsing; no specific sanitization or filtering for prompt injection patterns within the data is implemented.
[COMMAND_EXECUTION]: The skill performs filesystem operations to save generated reports. Specifically, the export_to_excel and generate_html_dashboard methods in SKILL.md write to the local disk. These are legitimate functions for a reporting tool and operate within the requested filesystem permission scope.

kpi-dashboard