The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to process external data from X (Twitter) posts, which represents a potential surface for indirect prompt injection. However, the instructions include proactive safety measures to mitigate this risk.
Ingestion points: The agent is instructed to read content from X URLs, threads, and trend data (SKILL.md).
Boundary markers: The skill explicitly directs the agent to 'Separate quoted claims from verified facts' and 'Do not present an X claim as independent confirmation'.
Capability inventory: The skill utilizes the clips_create and clips_get tools to generate and manage video assets based on external input.
Sanitization: The process requires building a reviewable brief and explicitly mandates user approval before purchasing, publishing, or spending credits.
[COMMAND_EXECUTION]: The skill references tool calls such as clips_create and clips_get. These appear to be standard platform-authorized tool invocations associated with the BlitzReels service and do not involve arbitrary shell command execution.

blitzreels-x-video