Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

For Agents — Argon Language Reference

This part answers: as an agent writing, debugging, or exploring Argon code, what do I need to know that the rest of the book doesn’t already give me?

The rest of the Argon book is a tutorial walkthrough for human readers. This part is dense, info-rich, agent-oriented reference — designed for fast lookup over linear reading. Each subpage covers one focused concept with a “this page answers:” one-liner at the top and dense cross-links to related pages.

When to read what

You want toGo to
Look up the syntax for any declaration formQuick Reference
Step-by-step recipes for common modeling jobsTask Recipes
Pick between alternatives (derive vs compute, which compute form, etc.)Decision Guides
Pick the right decidability tierTier Selection
Resolve a specific diagnosticError Recovery
Avoid an Argon-specific mistakeAnti-Patterns
Recognize a canonical Argon idiomIdioms
Integrate via CLI / MCP / IDE / skillsTooling
Look up a termGlossary
Look up a keywordAppendix A — Keyword Reference
Look up an operatorAppendix B — Operator Reference
Look up a diagnostic codeAppendix C — Diagnostic Codes
Understand the language end-to-endFoundations, Composition, Beyond

Scope

This part covers writing and reading Argon-the-language. It does not cover:

  • Working on the compiler / LSP / toolchain — that’s the AGENTS.md tree at the repo root and per-crate.
  • Why Argon was designed this way — that’s the RFDs at /rfd/.
  • What’s currently shipping or open — that’s GitHub Issues + Releases.
  • Compiler-internal API — that’s rustdoc on oxc.

Pages in this part