The Agent Skills Directory

[SAFE]: The skill defines templates for internal communications and utilizes standard tools for context gathering. No malicious instructions, unauthorized data access, or code execution patterns were found.
[PROMPT_INJECTION]: The skill involves ingesting data from external communication channels, which represents a surface for indirect prompt injection. Ingestion points: Content is retrieved from Slack messages, corporate emails, and Google Drive documents (specified in examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md). Boundary markers: The instructions lack explicit delimiters to separate ingested data from agent instructions. Capability inventory: Access to tools for reading company chats and documents. Sanitization: The skill does not implement validation or sanitization of the retrieved content.

internal-comms