The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the @reticlehq/server package from the author's official NPM registry. It also fetches additional documentation from https://docs.reticle.sh/llms.txt.
[COMMAND_EXECUTION]: The skill includes shell commands for environment initialization via npx @reticlehq/server@latest init.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from a live running application, which constitutes an untrusted input source.
Ingestion points: Computed styles, element metadata, and animation timing data retrieved from the running application through the reticle_inspect and reticle_animations functions.
Boundary markers: No specific delimiters or "ignore instructions" directives are used when processing the data extracted from the application.
Capability inventory: The skill utilizes observation tools including reticle_snapshot, reticle_inspect, reticle_animations, and reticle_feedback to process and report on the application state.
Sanitization: The instructions do not specify any sanitization or validation of the CSS properties or animation data before they are processed by the agent.

design-system-compliance