The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to execute several shell commands in Step 0, including clay whoami, clay workflows actions list, and clay workflows actions schema. These are used to identify the environment and valid action parameters within the Clay platform and are consistent with the skill's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it processes untrusted data from multiple sources to generate automated communication.
Ingestion points: Untrusted data enters the agent context from CRM campaign records, accounts, and contacts in Step 3, as well as from external people-sourcing and email-finding services in Step 6.
Boundary markers: The instructions do not specify the use of delimiters or "ignore embedded instructions" warnings when processing these inputs for copy generation in Step 8.
Capability inventory: The skill has significant capabilities, including writing new records to the user's CRM and enrolling contacts into a sequencer (Gong) as described in Step 10.
Sanitization: There is no mention of sanitizing or escaping the external content before it is used to generate personalized email drafts.
[EXTERNAL_DOWNLOADS]: In Step 6, the skill fetches data from external sourcing services (cpj-find-lists-of-people and enrow-find-work-email) to find new leads and their contact information. These are standard platform-integrated services used for the skill's primary purpose.

build-event-invite-list