Borrowing it
Nothing to install: this file belongs to dylansantwani/openbrowser. 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/dylansantwani/openbrowser/main/CLAUDE.mdgit clone --depth 1 https://github.com/dylansantwani/openbrowserWrote 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/dylansantwani/openbrowser/claude-md)<a href="https://agentmods.dev/instructions/dylansantwani/openbrowser/claude-md"><img src="https://agentmods.dev/badge/instructions/dylansantwani/openbrowser/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 | $0.12730 | $0.12730 |
| Opus 5 | $0.06365 | $0.06365 |
| Sonnet 5 | $0.02546 | $0.02546 |
| Haiku 4.5 | $0.01273 | $0.01273 |
Grade A, and why
openbrowser CLAUDE.md scanned grade A with 1 finding 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 yesterday.
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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
import { spawn } from 'node:child_process'; How it starts
The opening of the file, as written. The whole thing — 892 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenBrowser — working guidance
Read this before changing anything in openbrowser/. It exists because most of
what is non-obvious here was learned by driving the extension against real sites
and watching it fail in ways that are invisible from unit tests.
What this is
A Chrome extension plus a zero-dependency MCP server, letting any MCP client (Claude Code, opencode, Cursor) drive a real Chrome browser with the user's real logins.
MCP client ──stdio──> mcp-server ──ws://127.0.0.1:8848──> extension ──> tabs
Several MCP clients share one hub, and one hub drives several browsers. A session is bound to one browser and to one window inside it; the hub routes by that binding. See "The browser was never a singleton" below.
Two halves, two languages of failure:
mcp-server/is Node. Bugs here look like "the tool never appears in my client" or "every call times out".extension/is MV3. Bugs here look like "it says it worked but nothing happened".
Hard rules
Never add a dependency. Both package.json files have empty dependencies
and that is the single most important property of this project. npm install
failing is the most common reason a local MCP server does not work, and it fails
silently from the user's point of view. The WebSocket (src/ws.js) and the MCP
protocol (src/mcp.js) are hand-written for exactly this reason. If you need a
library, write the 200 lines instead.
stdout is protocol-only. In mcp-server/, anything written to stdout that
is not a JSON-RPC frame corrupts the stream and the client drops the connection
with no useful error. Human-readable output goes to stderr.
The service worker has no memory. MV3 kills it after ~30s idle and restarts
it constantly. Anything durable lives in chrome.storage; anything else is
rebuilt on demand. Every top-level statement in background/ must be safe to
run many times.
Errors are written for a model to recover from. "ref e12 no longer exists — the page changed; take a fresh snapshot" beats "Error: null". Put the
recovery action in the message. This is not politeness; it is the difference
between an agent retrying correctly and an agent giving up.
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.
- yesterday Changed · +112 lines · +1,927 tokens per session 260b2d1cdc02
- 2d ago Changed · +36 lines · +556 tokens per session 38d426c8b443
- 6d ago First seen · 744 lines · 10,247 tokens per session scan A d00484153412
openbrowser CLAUDE.md is an instructions file published in the GitHub repository dylansantwani/openbrowser (1 stars, last pushed 2d ago), licensed MIT. It adds 12,730 tokens to every session, about $0.0636 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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.
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.
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).
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).
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.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.