The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and executes the @useclawlink/cli package from the official NPM registry via npx. This is the primary tool for the skill's operation.
[COMMAND_EXECUTION]: Instructs the agent to execute shell commands using the ClawLink CLI to manage authentication and interact with the Facebook API.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill retrieves potentially attacker-controlled content from Facebook through actions such as facebook_get_comments, facebook_get_page_posts, and facebook_get_page_conversations (SKILL.md).
Boundary markers: No explicit boundary markers or instructions to ignore instructions embedded in the external data are present.
Capability inventory: The skill includes several write-access capabilities that could be abused if the agent follows instructions found in external data, such as facebook_create_post, facebook_send_message, facebook_delete_post, and facebook_update_post (SKILL.md).
Sanitization: There is no documented sanitization, validation, or filtering of the content retrieved from Facebook before it is processed by the agent.

facebook