The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill workflow instructs the agent to run npx @open-slide/cli init, which downloads and executes arbitrary code from an external package registry without version pinning or source verification.
[EXTERNAL_DOWNLOADS]: The skill uses pnpm install and pnpm up @open-slide/core to fetch dependencies. These packages are associated with an unverified repository (github.com/1weiho/open-slide) rather than a trusted organization or well-known service.
[COMMAND_EXECUTION]: The skill executes multiple shell commands (npx, pnpm dev, pnpm build, pnpm preview) in the local environment to initialize and manage the workspace.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data that could influence its behavior or the code it generates.
Ingestion points: topic input, DESIGN.md file, and user comments on UI elements.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing workflow.
Capability inventory: The skill possesses the ability to execute shell commands and modify local files.
Sanitization: No evidence of sanitization or validation of external content before it is processed or used in code generation.

open-slide-studio