The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the official @useclawlink/cli package from the NPM registry.
[COMMAND_EXECUTION]: The agent is instructed to run shell commands using the ClawLink CLI to manage authentication and interact with Miro board items.
[CREDENTIALS_UNSAFE]: The skill documents that authentication credentials are stored locally in the user's home directory at ~/.clawlink/credentials.json. This is standard practice for CLI-based authentication and not a security risk in this context.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it reads content from Miro boards (cards, sticky notes, etc.).
Ingestion points: Miro data is retrieved via npx @useclawlink/cli run miro commands.
Boundary markers: The instructions explicitly state to "read first" and "confirm with the user before any write."
Capability inventory: The skill can read/write Miro data and manage connections via the CLI.
Sanitization: Human-in-the-loop review is requested for all write operations.

miro