The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted, user-generated content from external sources (Weibo posts, comments, and topic discussions).
Ingestion points: Data enters the agent context through tools like weibo_web_search, weibo_web_post_comments, weibo_web_user_posts, and weibo_web_comment_replies (documented in SKILL.md and references/sandbase-api-map.md).
Boundary markers: The instructions do not explicitly define boundary markers or delimiters for the fetched Weibo content to prevent the agent from mistaking data for instructions.
Capability inventory: The skill has the capability to search, retrieve user profiles, and fetch deep comment threads via the sandbase_call_tool gateway.
Sanitization: There is no mention of sanitization or filtering of the external content before it is processed by the agent.

weibo-research