The Agent Skills Directory

[SAFE]: No malicious patterns, command executions, or network exfiltration were found. The skill is entirely composed of educational content and email templates.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface as it ingests data from external context files.
Ingestion points: The skill references dependencies such as icp-sales.md, value-prop.md, and proof-points.md from the sales-context skill, as well as lead lists from lead-generator.
Boundary markers: Absent. The email templates interpolate variables like [Produkt], [Obserwacja], and [Imię] directly into the email body without delimiters to prevent the agent from obeying embedded instructions.
Capability inventory: None identified. The skill does not perform subprocess calls, file-system writes, or network operations in the provided file.
Sanitization: Absent. There is no evidence of data sanitization or validation before interpolation into the generated email content.

cold-email