The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill utilizes dynamic discovery and resolution of platform routines and workflow actions by ID at runtime (e.g., clay routines get <id>). This involves loading metadata and schemas for execution units that are not statically defined in the skill, representing a dynamic loading pattern from computed paths.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from event registrant exports (CSV files) which are used to update CRM records and enroll contacts in sequences. This presents a vulnerability surface for malicious data to influence agent behavior.
Ingestion points: Registrant export CSV file processed in Step 2.
Boundary markers: The skill does not define specific prompt delimiters for the ingested data.
Capability inventory: Performs CRM record writes (Salesforce/HubSpot), executes platform CLI commands, and enrolls contacts in external sequencing tools (Outreach/Salesloft/Gong).
Sanitization: No specific sanitization or filtering of CSV content is mentioned; however, the skill enforces significant mitigations including mandatory human review of all CRM sync files (Step 3) and manual redlining of generated outreach copy (Step 8) before any live actions occur.
[COMMAND_EXECUTION]: The skill instructs the agent to execute clay CLI commands (whoami, routines list, workflows actions) to verify the workspace environment and discover available automation components. These commands are vendor-specific utilities for interacting with the Clay platform infrastructure.

event-follow-up-router