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/thronapple/chrome-local-mcp/agents-mdgit clone --depth 1 https://github.com/thronapple/chrome-local-mcpWrote 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/thronapple/chrome-local-mcp/agents-md)<a href="https://agentmods.dev/instructions/thronapple/chrome-local-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/thronapple/chrome-local-mcp/agents-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 | $0.00714 | $0.00714 |
| Opus 5 | $0.00357 | $0.00357 |
| Sonnet 5 | $0.00143 | $0.00143 |
| Haiku 4.5 | $0.00071 | $0.00071 |
Grade B, and why
chrome-local-mcp AGENTS.md scanned grade B with 2 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 3d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Edit `~/.codex/config.toml`: Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl http://localhost:9222/json/version # Verify Chrome CDP is reachable How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This file provides guidance to Codex (Codex.ai/code) when working with code in this repository.
Project Overview
chrome-local-mcp is a lightweight MCP (Model Context Protocol) server that connects to a Windows Chrome instance via CDP (Chrome DevTools Protocol). Designed for WSL environments where Codex needs to control a real browser with human-in-the-loop support for CAPTCHAs and verification challenges.
Build & Run
npm run build # Compile TypeScript → dist/
npm run dev # Watch mode
npm start # Run the MCP server (requires Chrome with --remote-debugging-port=9222)
Register with Codex
Codex mcp add --scope user chrome-local -- node /mnt/d/repository/chrome-local-mcp/dist/index.js
Register with Codex CLI
Edit ~/.codex/config.toml:
[mcp_servers.chrome-local]
command = ["node", "/mnt/d/repository/chrome-local-mcp/dist/index.js"]
Test
npm test # Quick tool listing test
curl http://localhost:9222/json/version # Verify Chrome CDP is reachable
Prerequisites
Windows Chrome must be running with remote debugging:
chrome.exe --remote-debugging-port=9222
Architecture
Codex (WSL) → stdio → MCP Server (WSL/Node.js) → CDP WebSocket → Chrome (Windows)
- Transport: stdio (MCP standard for CLI agents)
- CDP connection:
chrome-remote-interfacenpm package, single persistent connection with auto-reconnect - Token strategy: Long content saved to temp files (
/tmp/chrome-mcp/), only summaries returned to LLM context
Source Structure
src/index.ts— Entry point, CLI arg parsing, server bootstrapsrc/cdp.ts— CDP connection management (connect, reconnect, target switching)src/detect.ts— Shared challenge auto-detection (Cloudflare, CAPTCHA, login walls, etc.)src/tools/navigation.ts— navigate, wait_for, go_backsrc/tools/content.ts— get_content (smart extraction), evaluate (JS exec), screenshotsrc/tools/interaction.ts— click, fill, press_key, scrollsrc/tools/tabs.ts— tab_list, tab_open, tab_switch, tab_closesrc/tools/search.ts— Google search with auto CAPTCHA detectionsrc/tools/human.ts— wait_for_human, check_page_statussrc/types/chrome-remote-interface.d.ts— Type declarations for CDP client
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.
- 3d ago First seen · 74 lines · 714 tokens per session scan B 974c64fa8a0a
chrome-local-mcp AGENTS.md is an instructions file published in the GitHub repository thronapple/chrome-local-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 714 tokens to every session, about $0.0036 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
neobrowser AGENTS.md
Instructions for pitiflautico/neobrowser, covering agents.md — orientation for ai agents & contributors, build, run, test (rust — primary), how it works (rust module map), conventions and ci and release gates.
cdpilot CLAUDE.md
Instructions for mehmetnadir/cdpilot, covering cdpilot, kimlik, hızlı başlangıç, dosya haritası and dikkat edilecekler.
cdpilot GEMINI.md
Instructions for mehmetnadir/cdpilot, covering context-mode — mandatory routing rules, blocked commands — do not attempt these, curl / wget — blocked, inline http — blocked and webfetch / web browsing — blocked.
native-devtools-mcp AGENTS.md
AGENTS.md instructions for vectora-foundry/native-devtools-mcp, covering agent context: native-devtools-mcp, 🧠 core reasoning loop, 🗺️ capabilities matrix (strategy guide), 🛠️ tool definitions & schemas and 1. vision & perception (the "eyes").
Hover CLAUDE.md
Instructions for Hyperyond/Hover, covering directory guide, core documentation index, what hover is, direction (mcp-first, 2026-06) and workspace directories.
worbrow AGENTS.md
AGENTS.md instructions for noisystreet/worbrow, covering agents.md, project identity, hard constraints (revert on violation), modifying architecture docs and commit conventions.