The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill explicitly instructs the agent to execute remote code using npx -y @nomadamas/k-skill@0 instruct parking-lot-search. This command downloads and executes a package from the NPM registry to provide the primary instructions for the skill, which may vary from the static content provided in the skill files.
[EXTERNAL_DOWNLOADS]: The skill depends on external Node.js packages including @nomadamas/k-skill and parking-lot-search to perform its core functionality. It also directs the agent to fetch additional files and updates using the npx command.
[COMMAND_EXECUTION]: The skill requires the agent to run shell commands to manage its lifecycle, including npx -y @nomadamas/k-skill@0 files parking-lot-search and npx -y @nomadamas/k-skill@0 update.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input and data from external APIs, creating a potential surface for indirect prompt injection.
Ingestion points: User location strings, data.go.kr public parking API responses, and Kakao Map search results (instruction.md).
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for the external API data.
Capability inventory: Shell command execution via npx (SKILL.md), Node.js script execution (instruction.md), and network access to public APIs.
Sanitization: No sanitization, escaping, or validation logic is specified for the external content before it is processed or presented to the user.

parking-lot-search