The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data from restaurant searches, place details, and live availability queries which could contain untrusted content.
Ingestion points: External data enters the agent context through the stdout of search, place, and nearby commands executed via the HeyYumi CLI.
Boundary markers: The instructions do not specify explicit delimiters or guardrails for the agent when processing the JSON output from the tool.
Capability inventory: The skill has the capability to perform side-effect actions including reserve (sending reservation requests) and live (sending real-time inquiries to venue owners).
Sanitization: There is no mention of sanitization or filtering of the external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the CLI tool and fetch instructions at runtime.
Evidence: SKILL.md contains commands like npx -y @nomadamas/k-skill@0 instruct heyyumi and instruction.md uses npx -y heyyumi@0 for all operations.
[COMMAND_EXECUTION]: The skill relies on shell command execution to perform its primary functions, including searching and managing reservations.

heyyumi