The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to install the Vercel CLI and the @vercel/queue SDK. Both are official tools provided by Vercel, a well-known service provider.
[INDIRECT_PROMPT_INJECTION]: The skill defines patterns for processing queue messages, which represents a surface for ingesting untrusted data. This is the primary and intended function of the skill.
Ingestion points: Data enters the agent context through the handleCallback function (push mode) and the receive method (poll mode) as shown in references/quickstart.md and references/poll-mode.md.
Boundary markers: The provided code templates do not include specific delimiters or instructions to ignore embedded commands within the message payload.
Capability inventory: The skill facilitates queue operations including sending messages, receiving payloads, and acknowledging processing completion via the @vercel/queue SDK.
Sanitization: The reference implementation suggests direct processing of the message payload; application-level validation and sanitization are left to the developer.
[CREDENTIALS_UNSAFE]: The documentation correctly recommends using vc env pull to manage secrets via environment variables. This is a secure alternative to hardcoding credentials and aligns with standard industry practices.

vercel-queues