The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires downloading and installing external packages and source code to function.
Evidence in SKILL.md: References author-owned tools via npx -y @nomadamas/k-skill@0 instruct korean-cinema-search.
Evidence in instruction.md: Instructs the user or agent to run npm install -g daiso and git clone https://github.com/hmmhmmhm/daiso-mcp.git.
[REMOTE_CODE_EXECUTION]: The skill uses npx to download and execute code at runtime and suggests building external repositories.
Evidence in instruction.md: Multiple commands using npx --yes daiso are provided for searching theaters and movie times. It also includes a fallback workflow that involves cloning a repository, running npm install, and executing the resulting build with node dist/bin.js.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external cinema APIs, which represents a potential attack surface if the upstream data is compromised.
Ingestion points: Tool outputs from the daiso CLI, which fetches JSON data from various endpoints such as /api/cgv/theaters and /api/megabox/seats (documented in instruction.md).
Capability inventory: The skill has the capability to execute shell commands (npx, git, npm, node) and perform network requests as described in its operational workflow.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions that might be embedded in the retrieved cinema data.
Sanitization: There is no mention of sanitizing or validating the JSON data returned by the cinema APIs before the agent processes and summarizes it for the user.

korean-cinema-search