Skip to content

AGENTS.md

Latest commit

 

History

History
121 lines (88 loc) · 4.89 KB

AGENTS.md

File metadata and controls

121 lines (88 loc) · 4.89 KB

Repository Guidelines

Project Structure & Module Organization

This repository is a Next.js + TypeScript docs site.

  • app/: App Router pages, layouts, route handlers (page.tsx, layout.tsx, route.ts).
  • components/: Reusable UI and MDX-facing components.
  • content/docs/: Documentation source (.mdx) and section navigation via meta.json.
  • lib/, hooks/, contexts/, providers/, utils/: shared logic and state helpers.
  • scripts/: content generation utilities (OpenAPI markdown and llms.txt generation).
  • public/images/: static assets referenced by docs.

Build, Test, and Development Commands

Use Bun for all workflows.

  • bun install --frozen-lockfile: install dependencies consistently.
  • bun run dev: run local docs site on http://localhost:3030.
  • bun run build: generate docs artifacts, then run production Next.js build.
  • bun run start: serve the production build.
  • bun run lint or bun run check: run Biome checks across app/content/script directories.
  • bun run lint:fix or bun run check:fix: auto-apply Biome fixes.
  • bun run format: format code/content with Biome.
  • bun run validate-links: run link validation (lint.ts).
  • bun run generate-llms / bun run generate-openapi: regenerate derived documentation files.

Coding Style & Naming Conventions

  • TypeScript is strict (tsconfig.json); prefer explicit types for public helpers.
  • Formatting/linting is enforced by Biome (biome.json):
    • 2-space indentation, LF endings, 100-char line width
    • single quotes in JS/TS, double quotes in JSX
    • semicolons and trailing commas enabled
  • Follow existing file patterns: Next.js convention files in app/, kebab-case doc filenames in content/docs/.
  • Use the @/* path alias for internal imports where it improves clarity.

Testing Guidelines

There is no dedicated unit-test framework configured in this repo today. Treat these as required quality gates:

  1. bun run check
  2. bun run validate-links
  3. bun run build for changes that affect routing, generation, or site rendering

Commit & Pull Request Guidelines

History follows Conventional Commit style: feat:, fix:, chore:, ci: (example: fix: typo on docs page).

  • Keep commit subjects short, imperative, and scoped.
  • In PRs, include:
    • what changed and why
    • impacted paths (for example content/docs/overview/*)
    • screenshots/GIFs for UI or layout updates
    • linked issue/PR context when applicable

Security & Configuration Tips

  • Start from .env-example; never commit secrets.
  • Set LLMS_BASE_URL when generating llms.txt in non-default environments.

Writing Software

Behavioral guidelines to reduce common LLM coding mistakes. Merge with project-specific instructions as needed.

Tradeoff: These guidelines bias toward caution over speed. For trivial tasks, use judgment.

1. Think Before Coding

Don't assume. Don't hide confusion. Surface tradeoffs.

Before implementing:

  • State your assumptions explicitly. If uncertain, ask.
  • If multiple interpretations exist, present them - don't pick silently.
  • If a simpler approach exists, say so. Push back when warranted.
  • If something is unclear, stop. Name what's confusing. Ask.

2. Simplicity First

Minimum code that solves the problem. Nothing speculative.

  • No features beyond what was asked.
  • No abstractions for single-use code.
  • No "flexibility" or "configurability" that wasn't requested.
  • No error handling for impossible scenarios.
  • If you write 200 lines and it could be 50, rewrite it.

Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.

3. Surgical Changes

Touch only what you must. Clean up only your own mess.

When editing existing code:

  • Don't "improve" adjacent code, comments, or formatting.
  • Don't refactor things that aren't broken.
  • Match existing style, even if you'd do it differently.
  • If you notice unrelated dead code, mention it - don't delete it.

When your changes create orphans:

  • Remove imports/variables/functions that YOUR changes made unused.
  • Don't remove pre-existing dead code unless asked.

The test: Every changed line should trace directly to the user's request.

4. Goal-Driven Execution

Define success criteria. Loop until verified.

Transform tasks into verifiable goals:

  • "Add validation" → "Write tests for invalid inputs, then make them pass"
  • "Fix the bug" → "Write a test that reproduces it, then make it pass"
  • "Refactor X" → "Ensure tests pass before and after"

For multi-step tasks, state a brief plan:

1. [Step] → verify: [check]
2. [Step] → verify: [check]
3. [Step] → verify: [check]

Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.

These guidelines are working if: fewer unnecessary changes in diffs, fewer rewrites due to overcomplication, and clarifying questions come before implementation rather than after mistakes.