The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions create a surface for indirect prompt injection by directing the agent to ingest and summarize untrusted data from various communication tools.
Ingestion points: In examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md, the agent is instructed to read content from Slack, Google Drive, Email, and Calendar.
Boundary markers: Absent. The instructions do not specify the use of delimiters or provide directives to ignore embedded instructions found within the retrieved data.
Capability inventory: The agent is given broad read access to organizational communication platforms and is tasked with writing formatted summaries (newsletters, status updates, FAQs) intended for internal distribution.
Sanitization: Absent. No mechanisms for filtering, escaping, or validating content retrieved from external tools are mentioned in the skill instructions.

internal-comms