AGENTS.md

This file provides guidance to AI coding agents working in this repository.

Core rules

• Use yarn, not npm, for repo commands. This repo uses Yarn workspaces and Yarn 4.
• Run commands from the repo root unless a command explicitly says to run from a workspace.
• Never run bare tsc; use yarn typecheck from the repo root.
• Prefer targeted checks first. Avoid repo-wide test or e2e runs unless the change needs them.
• Keep changes scoped to the request and the affected package. Do not refactor unrelated code.
• Respect existing worktree changes. Do not revert user changes unless explicitly asked.
• Prefer editing existing files over creating new files. Do not add new documentation files unless requested.
• Use sentence case for headings, titles, labels, and documentation text.

Repo overview

This is the tldraw monorepo, an infinite canvas SDK for React applications. It is organized with Yarn workspaces.

Core packages:

• packages/editor - foundational infinite canvas editor with no default shapes, tools, or UI
• packages/tldraw - complete SDK with default UI, shapes, tools, and interactions
• packages/store - reactive client-side database, persistence, and migrations
• packages/tlschema - shape, binding, and record type definitions and validators
• packages/state - reactive signals library
• packages/sync and packages/sync-core - multiplayer sync packages
• packages/utils and packages/validate - shared utilities and validation helpers
• packages/assets - icons, fonts, translations, and bundled assets

Apps and examples:

• apps/examples - SDK examples and demos; the main place for example development
• apps/docs - documentation site at tldraw.dev
• apps/dotcom - tldraw.com app and workers
• apps/vscode - VS Code extension
• templates - starter templates for supported frameworks

Setup

Requires Node >=22.12.0. Enable Corepack before installing dependencies:

npm i -g corepack && yarn

Common commands

Development:

• yarn dev - start the examples app at localhost:5420
• yarn dev-app - start the tldraw.com client app
• yarn dev-docs - start the docs site
• yarn dev-vscode - start VS Code extension development
• yarn dev-template <template name> - run a template

Always run dev commands from the repo root. The root yarn dev runs each package's predev step, which generates build artifacts like packages/tldraw/tldraw.css. Running a per-workspace command (yarn workspace examples.tldraw.com dev) skips predev, so imports such as tldraw/tldraw.css fail to resolve. In a fresh git worktree, run yarn install first since worktrees start without node_modules.

Build:

• yarn build - build all changed packages incrementally
• yarn build-package - build SDK packages only
• yarn build-app - build the tldraw.com client app
• yarn build-docs - build the docs site

Testing:

• yarn test in a workspace - run tests in watch mode
• yarn test run in a workspace - run tests once
• yarn test run --grep "pattern" in a workspace - run matching tests
• yarn vitest - run all tests across the repo; slow, avoid unless necessary
• yarn e2e - run examples e2e tests
• yarn e2e-dotcom - run tldraw.com e2e tests

Code quality:

• yarn lint - lint the package or workspace
• yarn lint-current - lint changed files
• yarn typecheck - type check all packages and refresh assets
• yarn format - format the repo
• yarn format-current - format changed files
• yarn api-check - validate public API reports

Validation workflow

• For narrow package changes, run the relevant workspace test first, for example cd packages/tldraw && yarn test run --grep "SelectTool".
• For changes that affect shared types, migrations, editor behavior, or cross-package contracts, run yarn typecheck from the repo root.
• For public API changes, run yarn api-check and include intentional API report updates.
• For asset changes, run yarn refresh-assets or yarn typecheck so generated assets stay current.
• For docs changes, run the narrow docs checks or docs build only when the change affects generated content, MDX behavior, or site structure.
• For e2e behavior changes, run the smallest relevant e2e suite and update snapshots only when behavior intentionally changed.

Architecture notes

Reactive state:

• State is managed through @tldraw/state signals (Atom, Computed, and related primitives).
• Editor state is observable and dependency-tracked. Avoid bypassing existing reactive patterns.

Shapes:

• Shape behavior lives in ShapeUtil classes.
• Shape utils define geometry, rendering, handles, interactions, and SVG/export behavior.
• Add custom shape behavior through the established ShapeUtil patterns rather than one-off editor patches.

Tools:

• Tools are StateNode state machines.
• Complex tools use child states for pointer, keyboard, tick, and transition behavior.
• Keep interaction logic close to the tool state that owns it.

Bindings:

• Shape relationships use binding records and BindingUtil classes.
• Arrows and other connected shapes should update through binding utilities, not ad hoc shape mutation.

Managers:

• Editor subsystems live in packages/editor/src/lib/editor/managers/ as classes owned and disposed by the Editor.
• A manager that subscribes to events or holds a resource should extend EditorManager and register its cleanup so it runs on dispose(): addEditorEvent(event, fn) for editor bus events, register(fn) for everything else (store side effects, reactions, DOM listeners, child resources). Use editor.timers for timeouts/intervals/frames and editor.disposables for cleanup on the editor itself.
• Don't extend EditorManager for managers with no teardown. See the EditorManager doc comment for the full decision guide.

Store and schema:

• Store changes should respect migrations, validators, and schema versioning.
• Schema-affecting changes usually need updates in packages/tlschema and focused migration tests.

Where to work

• Use packages/editor for core editor primitives, geometry, managers, and UI-free behavior.
• Use packages/tldraw for default shapes, default tools, UI, and integration tests that need the full SDK.
• Use apps/examples for runnable SDK examples and demonstrations.
• Use apps/docs/content for documentation articles and release notes.
• Use apps/dotcom/client for tldraw.com frontend behavior.
• Use apps/dotcom/*-worker for Cloudflare worker behavior.
• Use templates for starter project changes.

Testing guidance

• Unit tests live alongside source files as *.test.ts.
• Integration tests commonly live in packages/tldraw/src/test/.
• E2E tests live in apps/examples/e2e/ and apps/dotcom/client/e2e/.
• Test in packages/tldraw when default shapes, tools, bindings, or UI are involved.
• Test in packages/editor for core editor behavior that should not depend on default shapes or UI.
• Prefer comparing whole objects in assertions when that gives a clearer failure than checking fields one by one.
• See skills/write-unit-tests/ and skills/write-e2e-tests/ for detailed test patterns.

Documentation and examples

• Docs live in apps/docs/content/.
• Examples live in apps/examples/src/examples/.
• Example folders use lowercase kebab-case names.
• Example README frontmatter drives the examples site; keep titles and descriptions sentence case.
• Update docs or examples when an API or user-facing behavior changes.
• See skills/write-docs/, skills/write-example/, and skills/write-release-notes/ for task-specific guidance.

Skills

• Canonical agent skills live in skills/.
• .agents/skills is a symlink to ../skills for generic agent compatibility.
• .claude/skills is a symlink to ../skills for Claude compatibility. Keep skills/ as the source of truth.
• .cursor/skills is a symlink to ../skills for Cursor compatibility.
• Skill folders use skill-name/SKILL.md with YAML frontmatter containing at least name and description.
• Put reusable scripts, references, and assets inside the relevant skill folder.
• Do not duplicate skill content for different agents; add compatibility pointers or symlinks instead.
• See skills/skill-creator/ before creating or restructuring skills.
• User-facing workflow skills include skills/pr/, skills/issue/, skills/take/, skills/commit-changes/, and skills/clean-copy/.

Code conventions

TypeScript:

• Follow existing file-local style and abstractions.
• Use workspace types and helpers rather than duplicating definitions.
• Keep public API changes deliberate and reflected in API reports.
• Avoid boolean or ambiguous positional options in new APIs when a named object or enum would make call sites clearer.

React and UI:

• Follow existing component patterns in the relevant app or package.
• Keep user-facing text concise and sentence case.
• Avoid broad UI rewrites when a focused component change is enough.

Generated files:

• Do not hand-edit generated assets, API reports, or schemas unless the repo already expects that file to be edited directly.
• Run the owning generator command when generated output needs to change.

Dependencies:

• Keep dependencies workspace-appropriate.
• If changing dependency manifests or lockfiles, make sure the lockfile update is intentional and included.

Writing style

• Use sentence case for Markdown headings, UI labels, docs titles, PR titles, and issue titles.
• Capitalize proper nouns, acronyms, and code names normally, for example PostgreSQL, WebSocket, and NodeShapeUtil.
• Use direct, concrete language.
• Do not include AI attribution in commits, PR descriptions, issues, docs, release notes, or generated written content.

Git and PR notes

• Keep commits focused when asked to commit.
• Use semantic PR titles for pull requests: <type>(<scope>): <description>.
• Never add yourself or an AI tool as a co-author.
• See skills/pr/ and skills/issue/ for GitHub workflows, and skills/write-pr/ and skills/write-issue/ for repository content standards.

Awesome Agent Conventions

A curated field guide to the convention files AI agents read, write, and act on.

22 conventions across 11 categories. From common project instruction files to newer agent-web discovery and trust formats.

Agent tools increasingly rely on plain files in a repository or website root: instructions, memory, rules, tool connections, prompt assets, discovery metadata, and protocol hints. The names are easy to mix up, and the adoption levels vary a lot.

This repo keeps the map practical:

• Know what a file is for. Each entry names the convention, usual filename, primary readers, and spec or source.
• Study real examples. Examples are fetched from public repositories by script, with provenance kept at the top of each file.
• Separate practice from proposal. Maturity labels show what is widely used, what is early, and what is still only proposed.

Contents

• Instruction & context · category page
  • 🟢 AGENTS.md
  • 🟢 CLAUDE.md
  • 🟢 Tool-specific instruction files
  • 🟠 OKF (Open Knowledge Format)
• Memory & state · category page
  • 🟢 MEMORY.md
  • 🟢 Memory Bank
• Spec-driven development · category page
  • 🟢 Spec Kit
  • 🟢 Kiro steering files
• Skills & prompt assets · category page
  • 🟢 SKILL.md
  • 🟢 Prompt asset files
  • 🟢 Claude Code commands
  • 🟢 Copilot prompt & instruction files
• Tooling & connections · category page
  • 🟢 MCP server config
• Rules & ignore files · category page
  • 🟢 Rules files
  • 🟢 AI ignore files
• Design · category page
  • 🟢 DESIGN.md
• Web & discoverability · category page
  • 🟢 llms.txt
  • 🟢 pricing.md
• Agent-web trust · category page
  • 🟠 auth.md
  • 🔵 ai.txt
• Identity & protocols · category page
  • 🟠 Agent Cards (A2A)
• Proposed namespace · category page
  • 🔵 The protocols.md namespace

What counts

This list is intentionally narrow. A file belongs here when it is a convention for agent behavior or agent-readable metadata: instructions, memory, skills, rules, tool config, prompt assets, or web-discovery hints.

Any file type can qualify - .md, .txt, .prompty, .json, dotfiles, or a directory pattern. Human-first project docs such as README.md, CONTRIBUTING.md, SECURITY.md, and CHANGELOG.md stay out unless the file has become an agent convention in its own right.

Maturity tiers

The badge is a claim about adoption, not quality. It keeps a proven convention from being presented the same way as a new idea.

Badge	Tier	Meaning
🟢	Adopted	Used in production by multiple tools, projects, or teams.
🟠	Emerging	Published by a real organization, but still early or limited in adoption.
🔵	Proposed	Publicly described, but without clear adoption beyond the proposal.

Instruction & context

Standalone page: categories/instruction-context.md

	Convention	Files	Read by	Spec
🟢	AGENTS.md	AGENTS.md	Most coding agents - OpenAI Codex, Cursor, Jules, Aider, Gemini CLI, Zed, and others	spec ↗
🟢	CLAUDE.md	CLAUDE.md	Claude Code, and tools that read the Claude memory convention	spec ↗
🟢	Tool-specific instruction files	GEMINI.md AGENT.md QWEN.md WARP.md CONVENTIONS.md copilot-instructions.md	Each file is read by its namesake tool - Gemini CLI, Amp, Qwen Code, Warp, Aider, GitHub Copilot - often alongside or as a bridge to AGENTS.md	spec ↗
🟠	OKF (Open Knowledge Format)	.md	Agents over MCP (okfy, openknowledge, superops okf CLIs); Google's knowledge-catalog ingests bundles	spec ↗

• AGENTS.md - A plain-Markdown "README for agents" - build/test commands, conventions, and gotchas an agent needs before touching the code. The most widely adopted cross-tool instruction file.
• CLAUDE.md - Anthropic's memory file for Claude Code - loaded automatically at session start to carry project commands, style rules, and standing instructions across turns.
• Tool-specific instruction files - Per-tool instruction files that predate or coexist with AGENTS.md. Some tools now default to AGENTS.md while keeping legacy filenames alive, so these variants still matter when auditing real repositories.
• OKF (Open Knowledge Format) - A machine-first organizational knowledge base: a version-controlled folder of typed Markdown files (one concept per file) that any agent reads as ground-truth context. Open-sourced by Google Cloud in 2026 as the content layer to MCP's transport.

Memory & state

Standalone page: categories/memory-state.md

	Convention	Files	Read by	Spec
🟢	MEMORY.md	MEMORY.md	Claude Code's auto-memory - the per-project MEMORY.md index it writes and re-reads each session	spec ↗
🟢	Memory Bank	projectbrief.md productContext.md activeContext.md systemPatterns.md techContext.md progress.md	Cline, Roo Code, and Cursor (via the Memory Bank custom-instructions pattern)	spec ↗

• MEMORY.md - A persistent, agent-maintained index of durable facts - written and re-read across sessions so an agent accumulates project memory instead of relearning each time.
• Memory Bank - Cline's structured memory system - a set of Markdown files an agent reads at the start of every task to reconstruct full project context after its session memory resets. The six files shown are Cline's set; tools like Roo Code use an overlapping but different variant.

Spec-driven development

Standalone page: categories/spec-driven-development.md

	Convention	Files	Read by	Spec
🟢	Spec Kit	constitution.md spec.md plan.md tasks.md	GitHub Spec Kit's slash-command agents (Copilot, Claude, Gemini, Cursor, and more)	spec ↗
🟢	Kiro steering files	product.md structure.md tech.md	AWS Kiro (steering files are largely Kiro-specific)	spec ↗

• Spec Kit - GitHub's spec-driven workflow - a constitution plus per-feature spec → plan → tasks files that drive an agent through structured, reviewable implementation.
• Kiro steering files - Kiro's always-on steering docs - product, structure, and tech files that give the agent persistent project context outside of any single spec.

Skills & prompt assets

Standalone page: categories/skills-prompt-assets.md

	Convention	Files	Read by	Spec
🟢	SKILL.md	SKILL.md	Claude Agent Skills, Claude Code, Amp, Agent Skills-compatible tools	spec ↗
🟢	Prompt asset files	.prompty .prompt system_prompt.txt	Prompty tooling, Azure AI / Semantic Kernel, and apps that load externalized prompts	spec ↗
🟢	Claude Code commands	.md	Claude Code - project .claude/commands/ and user ~/.claude/commands/	spec ↗
🟢	Copilot prompt & instruction files	.prompt.md .instructions.md	GitHub Copilot in VS Code / Copilot CLI	spec ↗

• SKILL.md - A self-contained, model-invoked capability file that tells an agent when to load a reusable procedure and how to execute it.
• Prompt asset files - Externalized prompt files - Prompty's YAML-front-mattered .prompty, plain .prompt templates, and system_prompt.txt - that pull the prompt out of source code so it can be versioned and edited on its own. Only .prompty has a formal spec (prompty.ai); .prompt and system_prompt.txt are ad-hoc externalized-prompt filenames.
• Claude Code commands - A Markdown file Claude Code exposes as a /slash-command - a reusable, version-controlled prompt workflow, with optional frontmatter (allowed-tools, model, argument-hint) and $ARGUMENTS and shell placeholders (@file references are a general Claude Code prompt feature, not command-specific). Now converging with Agent Skills, but still widely committed in its own right.
• Copilot prompt & instruction files - Modular, path-scoped Copilot context: *.instructions.md auto-attach to matching files via an applyTo glob, while *.prompt.md are reusable prompts you invoke by name - the granular cousins of a single .github/copilot-instructions.md.

Tooling & connections

Standalone page: categories/tooling-connections.md

	Convention	Files	Read by	Spec
🟢	MCP server config	.mcp.json	Claude Code, Cursor, VS Code / Copilot, and Claude Desktop - every MCP host reads the same mcpServers schema, though the filename and path differ per tool	spec ↗

• MCP server config - A JSON file that tells an agent which Model Context Protocol servers to launch and how (command, args, env) - making a project's tool and data integrations portable, shareable, and version-controlled across every MCP-capable client.

Rules & ignore files

Standalone page: categories/rules-ignore-files.md

	Convention	Files	Read by	Spec
🟢	Rules files	.cursorrules .mdc .clinerules .clinerules/ (pattern) .windsurfrules	Cursor (.cursorrules / .mdc), Cline (.clinerules/ and legacy .clinerules), Windsurf (.windsurfrules)	spec ↗
🟢	AI ignore files	.aiignore .cursorignore .codeiumignore .aiexclude	JetBrains Junie (.aiignore), Cursor (.cursorignore), Codeium/Windsurf (.codeiumignore)	spec ↗

• Rules files - Per-tool rule files that scope agent behavior - older single-file forms (.cursorrules, .clinerules, .windsurfrules) and newer directory-based, glob-scoped forms (.cursor/rules/.mdc, .clinerules/, .windsurf/rules/.md).
• AI ignore files - gitignore-syntax files that fence an AI agent out of paths - secrets, vendored code, generated output - so they're never sent to the model as context.

Design

Standalone page: categories/design.md

	Convention	Files	Read by	Spec
🟢	DESIGN.md	DESIGN.md	Google Stitch natively; and coding agents (e.g. Claude Code) when pointed at it as design context	spec ↗

• DESIGN.md - A structured, machine-readable design specification - tokens, components, and layout intent - that an agent reads to generate or keep UI consistent with an established system. Open-sourced by Google Labs in 2026 as a cross-tool draft spec.

Web & discoverability

Standalone page: categories/web-discoverability.md

	Convention	Files	Read by	Spec
🟢	llms.txt	llms.txt llms-full.txt (pattern)	Docs sites publish it for LLM tools and crawlers - though no major provider has confirmed reading it	spec ↗
🟢	pricing.md	pricing.md	Agents and LLM browsers fetching a clean, parse-able pricing page	spec ↗

• llms.txt - A proposed-turned-widely-published standard: a root-level Markdown file giving LLMs a curated, link-rich map of a site's docs. Published across hundreds of developer-docs sites - though whether the major LLM providers actually read it remains unproven.
• pricing.md - The Markdown twin of a pricing page - same URL with a .md suffix - so an agent gets structured plans and numbers instead of scraping marketing HTML. A concrete, shipping instance of the page.md pattern.

Agent-web trust

Standalone page: categories/agent-web-trust.md

	Convention	Files	Read by	Spec
🟠	auth.md	auth.md	Agents discovering how to authenticate to a service (early adopters)	spec ↗
🔵	ai.txt	ai.txt	AI training/data-mining crawlers that voluntarily honor AI usage preferences; crawler support is not yet reliable	spec ↗

• auth.md - A Markdown file that tells an agent how to authenticate with a service - discovery of auth endpoints and flows. Shipped by WorkOS as a real, working convention, but adoption beyond it is still early.
• ai.txt - A text file declaring machine-readable consent, licensing, or policy preferences for AI training and data-mining. Spawning popularized the deployed root-file pattern, and a 2026 Internet-Draft now proposes a well-known URI; adoption and crawler obedience are still thin, so it stays 🔵.

Identity & protocols

Standalone page: categories/identity-protocols.md

	Convention	Files	Read by	Spec
🟠	Agent Cards (A2A)	agent-card.json agent.json (pattern)	A2A-compatible agents discovering another agent's capabilities	spec ↗

• Agent Cards (A2A) - The Agent2Agent (A2A) capability card - a JSON document at a well-known path advertising an agent's skills, endpoints, and auth so other agents can discover and call it. Now a Linux Foundation project at v1.0; adoption is growing but early.

Proposed namespace

Standalone page: categories/proposed-namespace.md

	Convention	Files	Read by	Spec
🔵	The protocols.md namespace	proof.md	- (no demonstrated readers; aspirational)	spec ↗

• The protocols.md namespace - A single maintainer's pre-registered namespace of ~74 aspirational .md "protocols" (proof.md, signature.md, reputation.md, …) staked as Schelling points for a future agent web. Published concept, no demonstrated adoption - see the page for the audited, honest caveats.

Maintaining examples

Example files are fetched, not invented. The extractor pulls them from public sources, stores them under conventions/<slug>/examples/<source>/<filename>, and adds a line-1 provenance comment. The examples remain under their upstream owners' licenses and terms; see THIRD_PARTY_EXAMPLES.md before reusing them.

To refresh everything:

pip install -r scripts/requirements.txt
python scripts/extract.py          # fetch real files + rebuild each convention's README
python scripts/build_readme.py     # rebuild this README from scripts/targets.json

Re-running is idempotent. A missing target prints a miss and is skipped. Examples are representative samples: any file over 256 KB (for example, a multi-MB llms-full.txt) is truncated with a marker pointing back to the full source. scripts/targets.json remains the source of truth for conventions that have not been migrated yet; the skill-md pilot uses local convention metadata instead. Edit the relevant source and re-run both scripts.

Shortcut targets are available in the Makefile:

make verify          # schema + generated files + example provenance + links
make extract         # refetch public examples and rebuild generated docs
make license-report  # summarize upstream licenses for vendored examples

CI keeps the generated files and links honest. The verify workflow checks that generated docs match catalog metadata and migrated local metadata, and that every spec, example, and instance URL still resolves on each pull request and weekly. Run the same link check locally with python scripts/check_links.py.

Contributing

Read CONTRIBUTING.md. In short: an entry must pass the filter above and carry evidence for its maturity tier. Add sources to scripts/targets.json for non-migrated conventions, run the scripts, and open a PR. The skill-md pilot uses local convention metadata instead. Do not hand-write example files.

Before proposing adjacent standards, check WATCHLIST.md. Project direction lives in ROADMAP.md.

License

The curation, scripts, and original prose in this repository are MIT. Vendored example files remain under their upstream owners' licenses and terms; see THIRD_PARTY_EXAMPLES.md.