The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates fetching external content from the X platform, which introduces a surface area for indirect prompt injection where untrusted data could influence agent behavior.
Ingestion points: Fetches tweet content from api.x.com/2/tweets/search/recent as described in SKILL.md.
Boundary markers: The implementation examples do not use delimiters or explicit instructions to distinguish between fetched data and system prompts.
Capability inventory: The skill is equipped with network access through the requests library and authenticated write capabilities via requests_oauthlib to post content back to X.
Sanitization: The provided code snippets do not demonstrate sanitization or filtering of the external tweet content before it is processed for tasks like voice modeling.

x-api