The Agent Skills Directory

[SAFE]: The skill provides instructions for a manual review process of draft documents. It does not include any scripts, command executions, or network requests.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface as it instructs the agent to process content from drafts and a blueprint file.
Ingestion points: Data is ingested from blueprint/60-bandeja.md and draft messages.
Boundary markers: The instructions do not define specific delimiters to separate untrusted draft content from system instructions.
Capability inventory: The skill possesses no dangerous capabilities such as file writing, network access, or shell command execution.
Sanitization: There is no mention of sanitizing or escaping the content read from external files.

bandeja