The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface when processing external data from Moda canvases. It includes explicit instructions to treat such data as non-executable and to ignore any embedded directives.
Ingestion points: The skill reads data from canvases via moda canvas read (documented in references/capability-map.md).
Boundary markers: references/contract.md contains explicit instructions to treat canvas content as data and ignore embedded directives.
Capability inventory: The skill utilizes the Bash tool to execute moda CLI commands.
Sanitization: The skill provides an explicit policy for the agent to distinguish data from instructions during design tasks.
[EXTERNAL_DOWNLOADS]: The skill guides the user to install the @moda-design/moda CLI and associated skills using standard package managers. These resources belong to the vendor's own namespace.
[COMMAND_EXECUTION]: The skill invokes the moda CLI tool via Bash for operational tasks such as system health checks (moda doctor), account management, and workflow routing. These commands are constrained to the platform's command namespace.

moda-core