The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the @useclawlink/cli tool using npx to perform actions in Mailchimp.
Evidence: npx @useclawlink/cli actions mailchimp, npx @useclawlink/cli run mailchimp, etc., in SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill fetches the @useclawlink/cli package from the NPM registry at runtime. This package is owned by the skill's vendor.
Evidence: References to npx @useclawlink/cli and the NPM package link https://www.npmjs.com/package/@useclawlink/cli in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill processes data retrieved from Mailchimp (such as campaign content, member details, and list activities) which could contain untrusted instructions.
Ingestion points: Data returned by Mailchimp search and retrieval actions (e.g., mailchimp_search_member, mailchimp_get_list_activities, mailchimp_get_campaign) defined in the actions table in SKILL.md.
Boundary markers: None identified in the skill instructions to separate external data from agent instructions.
Capability inventory: Command execution via npx @useclawlink/cli allows both reading and writing (creating/updating/deleting lists and campaigns) as seen in the actions list in SKILL.md.
Sanitization: While the skill explicitly instructs the agent to "confirm with the user before any write," there are no technical sanitization or filtering requirements specified for the data retrieved from Mailchimp.

mailchimp