mcp-environtment-browser
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-environtment-browserLog into GCP and list my compute instances"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-environtment-browser
MCP server untuk credential vault + browser execution broker, dengan subscription per-tab.
Status: Phase 1 COMPLETE (2026-08-16)
Repo ini sudah menyelesaikan Phase 1 (Strategi A: Local-First + License
Server Only) — 9 phase implementation closed, 21 logical commits, 176 unit
tests PASS, 87.24% coverage, hermes verify --json → ok: true.
Metric | Value |
Phase completed | 9/9 (Phase 1-9 + bootstrap) |
Quality gates | 10/10 passed |
Unit tests | 176 passed, 1 skipped |
Coverage | 87.24% (gate ≥80%) |
MCP tools | 13 (per spec §6.4) |
MCP prompts | 3 (per spec §6.4.1) |
License server endpoints | 4 (per spec §6.5) |
Build | wheel 43KB generated |
Manual validation | 5 items pending owner (init/serve/DB counter/license gate/pause-resume) |
Lihat docs/spec/01_phase1_local_first_license_only/state.json
untuk SSOT fase + quality gates. Lihat docs/spec/01_phase1_local_first_license_only/spec.md
untuk latar belakang, scope, dan strategi implementasi fase 1.
Related MCP server: BrowserMCP Secure
Apa Ini (Goal)
Sebuah MCP server yang memungkinkan agent (Hermes, OpenAI, Claude Code, dll) menjalankan tugas di browser — seperti login ke TikTok, scrape data Google, akses GCP, atau ambil data dari Coretax DJP — tanpa agent harus pegang credential mentah.
Prinsip utama:
Credential vault — username/password/SSH/OAuth token/env var, disimpan terenkripsi di local user
Browser execution broker — agent minta "scrape akun TikTok X", MCP yang handle login + ambil data + return JSON
Per-tab billing — setiap
browser.new_page()dihitung 1 tab, hit server license DB; user tidak terdaftar = tidak bisa pakaiMulti-agent — agent apapun yang bisa connect MCP (stdio transport) bisa pakai, tanpa modifikasi
13 MCP Tools + 3 MCP Prompts — Tools untuk runtime action (credential + browser + DevTools + pause/resume + list sessions + web monitoring), Prompts untuk pattern guidance (OAuth confirmation + browser debug + human intervention)
Web monitoring companion — real-time screenshot tiap tab di
http://localhost:9876, click-to-act untuk user interveneCross-platform — dirancang jalan di Linux/Windows/macOS
Stack & Lisensi
Item | Pilihan |
Bahasa | Python 3.10+ |
Server | FastAPI (license DB + per-tab counter) |
Client | Click/Typer CLI + MCP stdio |
Browser | Playwright (Python) |
Credential storage | libsecret/Secret Service API (Linux), extensible ke Windows/macOS |
Lisensi | MIT |
Branch default |
|
🚨 Sebelum Kontribusi — WAJIB BACA
AGENTS.md— strict guidelines, anti-over-engineering, owner workflow styleQUICKREF.md— TL;DR orientasi 2 menitdocs/spec/01_phase1_local_first_license_only/spec.md— latar belakang, K1-K7, scope, strategi A
Struktur Docs
.
├── AGENTS.md # ← BACA PERTAMA
├── QUICKREF.md # ← BACA KEDUA (orientasi 2 menit)
├── README.md # file ini
├── LICENSE # MIT
└── docs/spec/01_phase1_local_first_license_only/
├── spec.md # Latar belakang, K1-K7, scope, strategi A
├── state.json # SSOT fase + quality gates
├── knowledge.md # Reuse catatan (cua-driver, libsecret, Playwright)
├── roadmap_migration.md # Strategi A → D → fase 3 (untuk agent masa depan)
└── refactor/
├── 00_overview.md # Quick split matrix + scope contract
├── 10_mcp_server.md # Detail MCP tool spec
├── 20_license_server.md # Detail server endpoint spec
├── 30_client_arch.md # Local-first client structure
├── 40_distribution.md # PyInstaller binary + install script
├── 90_decisions_log.md # CAPTURE K1-K7 — JANGAN DIHAPUS
└── 95_anti_overengineering_guard.md # ← BACA KETIGA (anti-over-engineer)Status Implementasi
Phase | Strategi | Status |
Phase 1 | A. Local-first + license-server-only | ✅ COMPLETE (2026-08-16) |
Phase 2 | D. Hybrid credential-from-server + local browser | 📋 ROADMAP |
Phase 3 | Multi-tenant + per-target rate limit | 📋 ROADMAP |
Lihat roadmap_migration.md
untuk transisi detail.
Quick Start (setelah install.sh)
# 1. Install (idempotent)
bash install.sh && bash install.sh
# 2. Start license server (di background)
mcp-env-browser-license-server serve --port 8765 &
# 3. Register admin + get API key (sekali)
curl -X POST http://localhost:8765/license/register \
-H "Content-Type: application/json" \
-d '{"email":"owner@mcp-env-browser.local","plan":"dev"}'
# 4. Init client config
mcp-env-browser init --server-url http://localhost:8765 --api-key <key>
# 5. Start MCP stdio + monitoring UI
mcp-env-browser serve
# → MCP stdio untuk agent
# → http://localhost:9876 untuk monitoring UIKontribusi
Sampai spec disetujui owner (yadenmustopa), JANGAN tulis kode implementasi.
Tambah/update docs saja.
Setelah spec di-approve, kontribusi mengikuti workflow di
docs/spec/01_phase1_local_first_license_only/refactor/00_overview.md.
Lisensi
MIT — lihat LICENSE.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
Alicense-qualityAmaintenanceMCP server for Meshbrow that gives AI agents full browser automation capabilities with stealth anti-detection, enabling natural language control of browser sessions, data extraction, and multi-browser fleets.MIT- Alicense-qualityCmaintenanceSecurity-hardened MCP server that gives AI assistants full control over your real browser session, supporting 36 tools for navigation, data extraction, monitoring, and more.MIT
- Alicense-qualityDmaintenanceMCP server that lets AI assistants browse the web through your real Chrome with your cookies, sessions, and fingerprint, avoiding bot detection and CAPTCHAs. Enables web browsing, content extraction, and multi-step workflows via persistent tabs.32MIT
- AlicenseAqualityCmaintenanceMCP server providing browser automation for AI agents, enabling actions like clicking and typing, structured data extraction, content validation, multi-step task execution, and memory enrichment from web pages.8109MIT
Related MCP Connectors
A paid remote MCP for AI agent browser approval MCP, built to return verdicts, receipts, usage logs,
A paid remote MCP for AI agent browser MCP session, built to return verdicts, receipts, usage logs,
A paid remote MCP for AI agent browser DevTools MCP, built to return verdicts, receipts, usage logs,
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/yadenmustopa/mcp-environtment-browser'
If you have feedback or need assistance with the MCP directory API, please join our Discord server