{"schemaVersion":"opencli.cli.v1","slug":"llm","name":"LLM","shortName":"llm","binaryName":"llm","maker":{"slug":"simon-willison","name":"Simon Willison","type":"individual","url":"https://simonwillison.net","officialPlatformMaker":false,"featuredBuilder":true},"category":"AI","description":"Prompting, local plugins, and structured llm outputs from the terminal. Built by Simon Willison. Supports structured output — good for scripts and agents.","tagline":"Prompting, local plugins, and structured llm outputs from the terminal.","install":{"packageManager":"pipx","command":"pipx install llm","packageName":null,"npmPackage":null,"brewFormula":"llm","brewCask":null,"crateName":null,"pypiPackage":"llm","goPackage":null,"dockerImage":null},"verify":{"command":"llm --help","signal":"llm responds locally; authenticate before real work."},"quickStart":"llm 'Explain this command: rg TODO src'","exampleWorkflow":["llm 'Explain this command: rg TODO src'"],"agent":{"readiness":{"label":"Great for agents","score":75,"reasons":["Structured output is available for parsing.","Supports non-interactive/scripted use.","Works well in CI or repeatable automation."],"guardrails":["Verify identity/account before running task commands.","Network access is required; avoid leaking secrets in logs.","Start read-only, then ask before mutations."]},"packUrl":"https://opencli.co/cli/llm/agent.md","packMarkdown":"# OpenCLI Agent Pack: LLM\n\nUse this when an AI agent needs to work with `llm`.\n\n## What this CLI is for\nPrompting, local plugins, and structured llm outputs from the terminal. Built by Simon Willison. Supports structured output — good for scripts and agents.\n\nBest for: prompting, local plugins, and structured llm outputs from the terminal.\n\n## Agent readiness\nGreat for agents (75/100)\n- Structured output is available for parsing.\n- Supports non-interactive/scripted use.\n- Works well in CI or repeatable automation.\n\n## Install\n```sh\npipx install llm\n```\n\n## Verify before real work\n```sh\nllm --help\n```\nExpected signal: llm responds locally; authenticate before real work.\n\n## Safe starting commands\n```sh\nllm --help\n```\n\n```sh\nllm 'Explain this command: rg TODO src'\n```\n\n## Guardrails for agents\n- Verify identity/account before running task commands.\n- Network access is required; avoid leaking secrets in logs.\n- Start read-only, then ask before mutations.\n\n## Suggested agent instruction\nYou may use LLM (`llm`) for prompting, local plugins, and structured llm outputs from the terminal.. First install it if missing, then run the verify command. Start with read-only or inspection commands. Summarize what you found before changing anything. Ask for confirmation before commands that mutate remote state, spend money, deploy, delete data, merge code, or expose secrets.\n\nSource: OpenCLI\n"},"fit":{"bestFor":"prompting, local plugins, and structured llm outputs from the terminal.","useThisIf":"You want AI models and inference you can script with structured output.","skipIf":"You need something that works offline or without an account.","whatHappensNext":"Run `llm 'Explain this command: rg TODO src'` and see what comes back."},"capabilities":{"agentFriendly":true,"supportsJsonOutput":true,"supportsNonInteractive":true,"supportsDryRun":false,"requiresAuth":true,"requiresNetwork":true,"ciFriendly":true,"localFirst":false,"destructivePotential":"medium"},"taxonomy":{"useCases":["Prompting","Local plugins","Structured LLM outputs"],"aliases":["llm cli"],"keywords":["prompting models plugins"],"tags":["builder","agent-friendly","json-output"]},"links":{"website":"https://github.com/simonw/llm","github":"https://github.com/simonw/llm","docs":"https://llm.datasette.io/","opencli":"https://opencli.co/cli/llm","markdown":"https://opencli.co/cli/llm/agent.md","json":"https://opencli.co/cli/llm.json"},"metrics":{"githubStars":12008,"latestRelease":"2026-05-17T01:38:18Z","license":"Apache-2.0","metricLabel":"Homebrew installs (30d)","metricValue":1725,"metricSource":"homebrew","metricAsOf":"2026-06-08T11:00:48.024Z"}}