llm

Original source

Imported from historical reading log.

  • Extracted via api.fxtwitter.com fallback, then read linked TIL directly: https://til.simonwillison.net/llms/llm-shebang
  • Simon Willison shows a neat pattern for using his llm CLI in a shebang line, turning plain-English files or YAML templates into executable scripts.
  • The more interesting part is not the toy prompt examples but the tool-enabled/template-enabled scripts: parameterized prompts, embedded functions, and lightweight agentic shells around LLM/tool workflows.
  • Newsletter angle: a crisp example of LLMs collapsing the boundary between prompt, script, and tiny executable tool surface.