The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it instructs the agent to process untrusted information from sources like Slack messages, email threads, and external press. 1. Ingestion points: Slack messages, Google Drive documents, Email threads, and external press articles (referenced in 'examples/company-newsletter.md' and 'examples/faq-answers.md'). 2. Boundary markers: No delimiters or specific 'ignore instructions' directives are defined for isolating the ingested content. 3. Capability inventory: The skill guides the agent's capability to read and summarize sensitive organizational communications and documents. 4. Sanitization: No content validation, filtering, or sanitization logic is specified.
[NO_CODE]: The skill consists exclusively of instructional Markdown files and documentation; it contains no executable code, scripts, or binary dependencies.

internal-comms