The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from various source platforms (tweets, threads, video transcripts, blog posts) and uses this data to generate content that can be published via tools.
Ingestion points: The skill accepts external content in any format as the primary input for repurposing.
Boundary markers: There are no explicit instructions for the agent to treat the source content as data-only or to ignore embedded instructions within that content.
Capability inventory: The skill has access to sensitive tools including publish, update_post, and delete_post.
Sanitization: The risk is mitigated by a mandatory human-in-the-loop approval step ('Approval card. Show... only after the user approves') before any publishing or modification action is taken.

linkedin-repurposer