Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/richardplg2/figma-annotations/claude-mdgit clone --depth 1 https://github.com/richardplg2/figma-annotationsWrote 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/richardplg2/figma-annotations/claude-md)<a href="https://agentmods.dev/instructions/richardplg2/figma-annotations/claude-md"><img src="https://agentmods.dev/badge/instructions/richardplg2/figma-annotations/claude-md.svg" alt="Measured on agentmods" 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 | $0.00551 | $0.00551 |
| Opus 5 | $0.00275 | $0.00275 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
figma-annotations 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Figma Annotations — VS Code Extension
A VS Code extension that connects to Figma via WebSocket relay, lets you annotate design frames alongside code, and exports structured prompts for AI-assisted implementation.
Architecture
src/
extension.ts — Entry point, command registrations
types.ts — Data models (Collection, DesignFrame, CodeAnnotation)
storageService.ts — Read/write .annotations.json at workspace root
collectionProvider.ts — TreeView provider (Collections -> Frames/Code -> Notes)
figmaService.ts — WebSocket client for Figma relay (port 3055)
figmaViewerPanel.ts — Webview panel for design frame viewer
rangeResolver.ts — Relocates code ranges when source files change
xmlFormatter.ts — Format collections as XML
markdownFormatter.ts — Format collections as markdown prompt for AI agents
webview/viewer.html — Figma viewer webview HTML
Key Concepts
- Collection: A named group of AnnotationItems (CodeAnnotation | DesignFrame)
- DesignFrame: A Figma frame with nodeId, cached image, node tree, and notes
- CodeAnnotation: A code selection with file path, range, and optional note
- DesignNote: A note attached to a specific node within a DesignFrame
Commands
| Command | Description |
|---|---|
createCollection |
Create a new named collection |
connectFigma |
Connect to Figma relay via channel ID |
loadSelection |
Import selected Figma frames into a collection |
addCodeToCollection |
Add editor selection as code annotation |
copyCollectionAsXml |
Copy collection as XML to clipboard |
insertCollectionToTerminal |
Insert XML into active terminal |
exportCollectionsAsPrompt |
Multi-select export as markdown prompt with MCP tools reference |
openFigmaViewer |
Open design frame in webview panel |
WebSocket Relay
The extension communicates with Figma through a WebSocket relay at ws://localhost:3055. This is the same relay used by cursor-talk-to-figma-mcp.
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.
- 5d ago First seen · 50 lines · 551 tokens per session scan A bc4199fb8ca5
figma-annotations CLAUDE.md is an instructions file published in the GitHub repository richardplg2/figma-annotations (11 stars, last pushed 4mo ago), licensed MIT. It adds 551 tokens to every session, about $0.0028 per session on Opus 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
figwright AGENTS.md
AGENTS.md instructions for awdr74100/figwright, covering agents.md, architecture, layout, tech stack and commands.
figwright CLAUDE.md
Claude Code instructions for awdr74100/figwright: Read AGENTS.md first — it's the single source of guidance for this repo: architecture, layout, tech stack, commands, conventions, release flow, and gotchas.
UI-perfect CLAUDE.md
Claude Code instructions for nv-minh/UI-perfect, covering ui-perfect / specgate, cấu trúc, quy tắc quan trọng, lệnh hay dùng and figma file mẫu.
Aurelius CLAUDE.md
Claude Code instructions for PMDevSolutions/Aurelius, covering claude.md, project overview, project structure, development scripts and lint and format code.
Figma-to-Code_MCP CLAUDE.md
Instructions for vincegx/Figma-to-Code_MCP, covering claude.md, project overview, essential commands, docker environment (primary) and start the application (hot reload enabled).
kiro-for-codex AGENTS.md
Instructions for atman-33/kiro-for-codex, covering agents.md — kiro for codex, steering documents, decision precedence, agent behavior contract and paths & i/o.