The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs runtime performance audits using the command npx motionscore <url> --agent, which executes a CLI tool against user-provided URLs. It also handles library migration by managing package installation and uninstallation.- [DYNAMIC_EXECUTION]: The skill features interactive components like open-transition-editor that provide dynamic UI and feedback within the chat, allowing the agent to receive and apply real-time configuration changes.- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes external content from documentation servers and example repositories, which could potentially contain malicious instructions or biased content.- [INDIRECT_PROMPT_INJECTION]: Ingestion points: Content is ingested from external resources via search-motion-docs, search-motion-source, and the results of runtime audits performed on arbitrary URLs.- [INDIRECT_PROMPT_INJECTION]: Boundary markers: There are no explicit instructions or delimiters defined to isolate external documentation or source code from the agent's primary instruction set.- [INDIRECT_PROMPT_INJECTION]: Capability inventory: The skill possesses capabilities for shell command execution, complex file modification, and extensive external data retrieval.- [INDIRECT_PROMPT_INJECTION]: Sanitization: No specific sanitization or validation logic is implemented for the documentation or code examples retrieved from external providers before being processed by the agent.

motion