The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is designed to access and summarize highly sensitive corporate information from sources including Slack channels, private emails, Google Drive documents, and Calendar events.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Slack, Email, and Google Drive, creating a vulnerability surface for indirect prompt injection attacks where malicious instructions hidden in messages or documents could hijack the agent's behavior. Ingestion points: Slack messages, emails, and Google Drive document content referenced in examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md. Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the ingested data. Capability inventory: The skill utilizes the agent's integration with communication and document platforms to gather data. Sanitization: No sanitization, validation, or escaping logic is applied to the retrieved external content before it is processed by the agent.

internal-comms