The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes code designed to process external lead data, which creates a surface for indirect prompt injection.
Ingestion points: The rows array parameter in src/clay/cost-filter.ts and src/clay/sampler.ts accepts untrusted data input.
Boundary markers: No delimiters or specific instructions to ignore embedded prompts are provided in the scripts.
Capability inventory: The skill utilizes Read, Write, Edit, and Grep tools; the provided scripts do not perform network operations or subprocess execution.
Sanitization: The scripts implement business logic filtering (e.g., domain checks) but do not perform security-focused sanitization of the input strings.
[EXTERNAL_DOWNLOADS]: The skill references official documentation and pricing pages from Clay for user reference. These references target a well-known service and do not involve the download or execution of remote code.

clay-cost-tuning