The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation mentions an installation command for a related skill (sales-do) using npx from the vendor's repository (sales-skills/sales). This is a standard extension mechanism within the vendor's ecosystem.
[COMMAND_EXECUTION]: The skill provides templates for curl commands to interact with Zapier webhooks and an npx command to add related functionality. These are presented as educational examples for the user to implement.
[CREDENTIALS_UNSAFE]: The documentation discusses the Ruzuku API Key and API Secret required for Zapier authentication. No actual credentials are hardcoded; only placeholders and instructions on how to securely retrieve them from the Ruzuku dashboard are provided.
[PROMPT_INJECTION]: The skill implements an indirect prompt injection surface through its "learnings" mechanism. It reads from and appends to references/learnings.md to persist platform knowledge across sessions.
Ingestion points: Reads references/learnings.md at the beginning of Step 1 to incorporate previous platform knowledge.
Boundary markers: The instructions do not specify markers to distinguish stored learnings from system prompts.
Capability inventory: The skill provides instructions for generating curl and npx commands.
Sanitization: There is no explicit sanitization of user-provided "learnings" before they are written to the references/learnings.md file.

sales-ruzuku