The Agent Skills Directory

[SAFE]: The skill operates as described, providing a framework for sales forecasting using standard data processing techniques. It does not attempt to access sensitive system files or establish unauthorized network connections.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests and processes external data (CRM CSV exports or manually pasted deal lists).
Ingestion points: Data enters the agent context through the @$1 file reference (SKILL.md) and is read via the Read tool.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the processed data.
Capability inventory: The skill is limited to the Read, Glob, and Grep tools, which are used to find and parse pipeline information.
Sanitization: There is no evidence of input validation or sanitization to prevent data-embedded instructions from influencing the agent.

forecast