Figwright is a two-way Model Context Protocol server that lets coding agents exchange information with Figma through a companion plugin, turning designs into framework-aware code and editing the Figma canvas from code. Developers use it with agents such as Claude Code, Cursor, or Codex to read design context and create or modify Figma elements, and the catalogue contains skills and other workflow components for those interactions.
Borrowing it
Nothing to install: this file belongs to awdr74100/figwright. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/awdr74100/figwright/main/CLAUDE.mdgit clone --depth 1 https://github.com/awdr74100/figwrightWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/awdr74100/figwright/claude-md)<a href="https://agentmods.dev/instructions/awdr74100/figwright/claude-md"><img src="https://agentmods.dev/badge/instructions/awdr74100/figwright/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/awdr74100/figwright/claude-md"><img src="https://agentmods.dev/badge/instructions/awdr74100/figwright/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00720 | $0.00720 |
| Opus 5.5 | $0.00288 | $0.00288 |
| Sonnet 5.5 | $0.00144 | $0.00144 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
figwright CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
CLAUDE.md
Read AGENTS.md first — it's the single source of guidance for this repo: architecture, layout, tech stack, commands, conventions, release flow, and gotchas.
Claude Code specifics:
- This repo is a Figma MCP server. The project-scoped
.mcp.jsonlaunches the local build (packages/mcp/dist/index.mjs), so after changingpackages/mcporpackages/sharedyou mustpnpm buildand restart the MCP connection before thefigwrighttools reflect your change. - There are no git hooks; CI is the gate. When you edit through Claude Code, the PostToolUse hook (
.claude/hooks/format-on-edit.mjs) auto-formats and lints the file — don't hand-format. - The canonical checks before pushing are
pnpm typecheck && pnpm lint && pnpm format:check && pnpm knip && pnpm build && pnpm test(the same gates CI runs). - The MCP SDK is v2 —
@modelcontextprotocol/server(plus its transitive@modelcontextprotocol/core), not the v1@modelcontextprotocol/sdk.src/index.tsbuilds the server in a factory and hands it toserveStdio, which owns the transport and the protocol-era decision. - The SDK is a runtime dependency — it generates every tool's JSON Schema and negotiates the protocol version — so a release can move what clients see while
tscstays green.packages/mcp/test/e2e/mcp-wire.test.tsis the gate for exactly that: it spawns the builtdistover real stdio, speaks raw JSON-RPC (never the SDK's ownClient, which would move in lockstep and hide the regression), and asserts the advertised tool set, each schema against an SDK-independent derivation, the dialect, annotations, prompts, a livetools/call, that a 2024-11-05 client is still served, that a 2026-07-28 client gets the modern era and is refused thepingthat revision deleted, and that the process still exits cleanly when the client goes away mid-session. It runs inpnpm testbut needspnpm buildfirst, and skips whendistis missing. - The
mcp-sdk-auditskill (.claude/skills/) owns SDK upgrades. The gate answers did anything break; the skill'sprobe.mjsanswers what moved, by snapshotting the live wire contract before and after a bump and diffing. Reach for it on any@modelcontextprotocol/*bump. - The
figma-typings-auditskill (.claude/skills/) owns@figma/plugin-typingsupgrades end to end: it diffs the.d.tsbetween the installed and target version (that package ships no changelog), maps the result onto the sandbox handlers, the hand-written Zod mirrors inshared, and the tool registry, then bumps and implements what's worth having. It diffs before upgrading — the installed version is the baseline. Reach for it on a Renovate bump PR for that package: a green pipeline does not mean the update was absorbed, sincesharedmirrors Figma's shapes with no compile-time coupling to the typings.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 14 lines · 720 tokens per session scan A 3dd3f81a1a19
figwright CLAUDE.md is an instructions file published in the GitHub repository awdr74100/figwright (1,188 stars, last pushed yesterday), licensed MIT. It adds 720 tokens to every session, about $0.0029 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.