The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references several official Scalar domains for documentation, dashboards, and schemas.
Links include docs.scalar.com, dashboard.scalar.com, and registry.scalar.com.
It also references the official starter repository at github.com/scalar/starter.
[COMMAND_EXECUTION]: The skill documents the use of the @scalar/cli tool via npx for various project management tasks.
Commands include project init, check-config, preview, publish, and upgrade.
These are standard operations for managing Scalar Docs projects as described in the documentation.
[INDIRECT_PROMPT_INJECTION]: The configuration schema described in the skill includes a siteConfig.head property that allows for the injection of custom scripts, styles, and meta tags.
Ingestion points: The agent receives instructions or data from the user to generate or modify the scalar.config.json file.
Boundary markers: None present in the documentation.
Capability inventory: The agent writes or modifies JSON files based on the schema.
Sanitization: The skill does not provide specific sanitization instructions for user-provided script or style content. While this is a documented feature of the Scalar Docs platform, it represents a potential surface for cross-site scripting (XSS) if the agent is directed to include untrusted content in these fields.

scalar-docs