The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes zapier-sdk commands to retrieve workflow metadata and user profile information. These commands are part of the Zapier SDK experimental surface and are used according to the skill's stated purpose.- [DATA_EXFILTRATION]: The skill accesses trigger_url and webhook_url fields. The instructions explicitly identify trigger_url as account-sensitive and direct the agent to avoid printing it unless specifically requested by the user.- [PROMPT_INJECTION]: The skill processes untrusted data from the Zapier API via the SDK, creating a surface for indirect prompt injection.
Ingestion points: JSON output from list-workflows, list-workflow-drafts, and list-workflow-runs.
Boundary markers: No specific delimiters or boundary markers are defined to isolate the processed data.
Capability inventory: The skill can execute CLI commands (zapier-sdk) and access workflow metadata.
Sanitization: There is no evidence of filtering or sanitization of the JSON data before it is presented to the agent.

workflows-list