The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides installation commands that download and execute code directly from the internet via shell piping. This pattern is found for the belt CLI, the uv package manager, and Node.js version managers.\n
Evidence: curl -fsSL https://cli.inference.sh | sh in SKILL.md and references/cli.md.\n
Evidence: curl -LsSf https://astral.sh/uv/install.sh | sh in references/cli.md.\n
Evidence: curl -fsSL https://fnm.vercel.app/install | bash in references/cli.md.\n
Evidence: curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.0/install.sh | bash in references/cli.md.\n- [PRIVILEGE_ESCALATION]: The skill includes a Windows installation command for the uv tool that uses the ExecutionPolicy Bypass flag to circumvent system-level security restrictions on script execution.\n
Evidence: powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex" in references/cli.md.\n- [COMMAND_EXECUTION]: The skill directs the agent to perform multiple system-level operations using the belt CLI and Node.js package runners to scaffold, test, and deploy applications.\n
Evidence: npx skills add belt-sh/cli in SKILL.md.\n
Evidence: Commands such as belt app init, belt app test, and belt app deploy are used throughout the skill instructions.\n- [EXTERNAL_DOWNLOADS]: The skill depends on numerous third-party libraries and tools downloaded from external repositories and registries during the application lifecycle.\n
Evidence: References to Node.js packages including zod, sharp, and @inferencesh/app.\n
Evidence: References to Python packages including torch, transformers, accelerate, and inferencesh.

building-inferencesh-apps