The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes content from Google Docs, which serves as an ingestion point for untrusted data that could influence agent behavior.
Ingestion points: The get action in skill-router.json returns the full document body via the body field ($response.body.content).
Boundary markers: No specific delimiters or "ignore instructions" warnings are applied to the retrieved document content.
Capability inventory: The skill has the ability to modify documents (append-text, replace-text) and share them (share), which could be targeted by instructions embedded in a document.
Sanitization: No sanitization or filtering is performed on the document content before it is passed to the agent's context.
[SAFE]: The skill interacts with well-known Google API endpoints (docs.googleapis.com and www.googleapis.com) to perform its functions. These communications are necessary for the skill's primary purpose and use official service domains.

documents