surf-mcp: Instructions file for Claude Code

CLAUDE.md

surf-mcp CLAUDE.md is an instructions file for Claude Code from shanevcantwell/surf-mcp. It costs 1,824 tokens per session, scanned A, original, MIT.

Instructions for Surf MCP, a server for controlling a visual web browser through an AI model. It describes visual grounding, where the model finds page elements from what they look like instead of relying on fixed CSS selectors.

In plain words
What is it for?
Use it when developing or operating Surf MCP, including visual element selection, Fara model actions, tool-call handling, and browser automation workflows.
Why use it?
It helps browser automation keep working when page markup or selectors are brittle. The instructions also define how Surf executes the model’s actions, such as clicking, typing, scrolling, visiting URLs, and pressing keys.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md.

This is shanevcantwell/surf-mcp's own configuration. It tells Claude Code how to work on surf-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything surf-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to shanevcantwell/surf-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/shanevcantwell/surf-mcp/release/v0.5.0/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/shanevcantwell/surf-mcp

Made for: Claude Code.

Wrote 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.

agentmods badge for surf-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/shanevcantwell/surf-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/shanevcantwell/surf-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/shanevcantwell/surf-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/shanevcantwell/surf-mcp/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.

agentmods 80×15 button for surf-mcp CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/shanevcantwell/surf-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/shanevcantwell/surf-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,824 This file is loaded in full into every session.
When invoked 1,824 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01824 $0.01824
Opus 5 $0.00912 $0.00912
Sonnet 5 $0.00365 $0.00365
Haiku 4.5 $0.00182 $0.00182

Measured 8d ago against content hash b5a508396573, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

surf-mcp 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 8d 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.

CLAUDE.md · 212 lines

How it starts

The opening of the file, as written. The whole thing — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Surf MCP Server

Purpose: MCP server for visual browser automation via Fara.

Version: 0.5.0


Core Concepts

Visual Grounding

Surf uses multimodal LLMs (Fara-7B via LM Studio, or Gemini/GPT-4V) to locate UI elements by natural language description instead of brittle CSS selectors.

An AI that can see the page doesn't need to parse HTML.

Direct Fara Execution (ADR-005)

Fara returns complete tool_calls, not just coordinates. We execute what Fara decides:

{
  "name": "computer_use",
  "arguments": {
    "action": "left_click",
    "coordinate": [624, 280],
    "reasoning": "The search button is a blue element..."
  }
}

Available actions: left_click, double_click, type, scroll, key, visit_url, terminate, wait

FaraToolCall Data Model

The FaraToolCall dataclass preserves Fara's full action context:

  • action: Action type (left_click, type, scroll, etc.)
  • coordinate: (x, y) pixel coordinates
  • text: Text to type (for type action)
  • direction: Scroll direction (up/down)
  • keys: Keys to press (for key action)
  • confidence: Model confidence (0.0-1.0)
  • reasoning: Fara's chain-of-thought explanation

Multi-Server LM Studio Support

Supports multiple LM Studio instances across different GPUs/machines:

  • Server Discovery: Probes each server's /v1/models manifest
  • Prefer Loaded: Prioritizes servers with Fara already loaded in VRAM
  • Fallback: Sequential retry across servers on failure

Configure in .env:

LMSTUDIO_SERVERS="rtx3090=http://localhost:1234/v1,rtx8000=http://192.168.137.2:1234/v1"
FARA_MODEL_IDS="microsoft_fara-7b,fara-7b-gguf,gao-zijian/fara-7b"
FARA_MAX_FAILURES=2
FARA_PROBE_TIMEOUT=2.0

Installation & Running

Docker (Recommended)

docker compose up

Direct Installation (Development)

pip install -e ".[dev]"
playwright install chromium
surf-mcp  # Run the server

Development Directives

Testing

# Setup
pip install -e ".[dev]"
docker build --target prod -t surf-mcp .

# Run tests
pytest                           # All tests (skips unavailable)
pytest -m "not llm"              # Skip LLM tests (for CI)
pytest -m docker                 # Docker container tests
pytest -m "docker and browser"   # Docker + browser tests
pytest -m llm -v -s              # Real LLM tests (requires LM Studio)
mypy src/

Read the full file on GitHub · 212 lines

Changes

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.

  1. 8d ago First seen · 212 lines · 1,824 tokens per session scan A b5a508396573

Subscribe to this mod's changes

surf-mcp CLAUDE.md is an instructions file published in the GitHub repository shanevcantwell/surf-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 1,824 tokens to every session, about $0.0091 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 2026-08-31.

Related

Other instructions, from other repositories

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.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

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).

microsoft/vscode · 6,785 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

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.

langchain-ai/langchain · 4,469 tokens

deepseek-harness AGENTS.md

AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.

deepseek-ai/deepseek-harness · 3,735 tokens