The Agent Skills Directory

[SAFE]: The skill instructions define a legitimate collaborative writing process using standard tools like create_file and str_replace. No suspicious command execution or exfiltration patterns were detected.
[PROMPT_INJECTION]: The skill exhibits an inherent surface for indirect prompt injection as it is designed to ingest and process data from external sources such as documents, files, and messaging integrations. This surface is necessary for its primary purpose of document creation. • Ingestion points: Reads from user-provided files, shared document links, and integrations like Slack, Microsoft Teams, and Google Drive (as described in SKILL.md). • Boundary markers: The instructions do not define specific delimiters or security warnings to wrap external content during ingestion. • Capability inventory: The agent has the ability to create files, modify files using string replacement, and invoke independent sub-agents for testing. • Sanitization: No sanitization or validation logic is specified for the external data being processed.

doc-coauthoring