mikita12/cognitive-browser

このページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。原文を見る

0 stars 0 forks 品質 81 トレンド 14MIT

Fast, token-efficient browser automation MCP server. Resolves routine web interactions via local DeBERTa-v3 NLI, escalating to LLMs only on ambiguity.

概要

Cognitive Browser enables AI coding assistants (Antigravity, Claude, Cursor) to navigate the web autonomously with consumed during DOM exploration, filling inputs, resolving CAPTCHAs, and clicking multi-hop routes. - ⚡ Navigates multi-hop websites using local ModernBERT / DeBERTa Cross-Encoder models in milliseconds (~3.3ms per element). - 🧠 Semantically binds arbitrary JSON form dictionaries (names, emails, resumes, CVs, filters) to DOM inputs without hardcoded CSS selectors. - 🚀 Automatically learns direct shortcut routes to pages and actions, delivering near-instant sub-15ms repeat interactions. - 🛡️ Restricts navigation to public destinations, blocks cloud metadata APIs (169.254.169.254), and protects against DNS rebinding. - 🍪 Seamlessly imports authenticated sessions from your local Firefox profile or cookie files to bypass Cloudflare and auth walls. - 🧩 Built with decoupled single-responsibility modules (parser, payload, captcha, cache, reflex, orchestrator, browser).

README

以下は原文 README の抜粋です。原文を見る。

Cognitive Browser

Zero-token autonomous browser agent & MCP server powered by pluggable local GPU Cross-Encoder / NLI routing (ReflexCore), universal cross-lingual form binding (PayloadBinder), Stealth Playwright (BrowserEngine), and self-healing RouteCache.

*Note on “Zero-Token”: Refers strictly to the Cognitive Browser MCP server and ReflexCore engine, which perform DOM candidate ranking, filtering, and autonomous multi-hop navigation locally on GPU with 0 LLM tokens. The orchestrating host agent (e.g. Ornith 1.5) still consumes LLM tokens (mean: 41,668 tokens/episode in Arm A vs 85,332 in Arm B).

Cognitive Browser enables AI coding assistants (Antigravity, Claude Desktop, Cursor) to navigate the web, search dynamic SPAs, and fill complex multi-field forms with 0 LLM tokens consumed by the browser server during DOM exploration and action execution.

Highlights

⚡ Zero-Token Autonomous Navigation & 51% Token Reduction: Navigates multi-hop websites on GPU in milliseconds (~3.3ms per candidate; 400+ DOM elements evaluated in <25ms), reducing overall agent LLM token consumption by 51.2% compared to standard Playwright MCP.
🔀 Hybrid Escalation Router: Executes confident navigation steps autonomously on GPU; when ambiguity is encountered, it seamlessly returns structured fallback candidates (partial_needs_fallback) for targeted host agent intervention via smart_action.
🔌 Pluggable Neural Models (Rerankers & NLI): Works out-of-the-box with any HuggingFace AutoModelForSequenceClassification — including multilingual Cross-Encoder Rerankers (mmarco-mMiniLMv2, bge-reranker-v2-m3, gte-reranker-modernbert-base) and NLI models (DeBERTa-v3, mDeBERTa-v3, ModernBERT-nli).
🧠 Dictionary-Free Cross-Lingual PayloadBinder: Partitions DOM controls by structural role (textbox, combobox, checkbox, upload_file) and binds arbitrary JSON payloads across languages (e.g., English keys to Polish forms or vice versa) with zero hardcoded synonym tables.
🚀 Self-Healing RouteCache: Learns deterministic shortcut routes in SQLite for near-instant repeat execution and automatically invalidates stale entries if a site’s DOM changes.
🛡️ SSRF & Security Guards: Restricts navigation to public destinations, blocks loopback/private networks and cloud metadata APIs (169.254.169.254), and mitigates DNS rebinding (per-request validation; resolve-then-connect TOCTOU remains).
🍪 Opt-In Cookie Sync & CAPTCHA Resolution: Can import authenticated sessions from local Firefox profiles (disabled by default; enabled via --enable-cookies flag) and integrates automated Turnstile / reCAPTCHA resolution.
🧩 Strongly-Typed Clean Architecture: Built with decoupled single-responsibility modules (models, parser, payload, captcha, cache, reflex, orchestrator, browser, security).

Architecture Overview

GitHub で README 全文を見る

インストール

This server does not publish a one-line install command.

Open the repository installation guide MCP 初心者？mcpskill.org のガイドから始めましょう →

設定

{
  "mcpServers": {
    "cognitive-browser": {
      "command": "python3",
      "args": ["-m", "cognitive_browser.mcp_server"],
      "cwd": "/path/to/cognitive-browser",
      "env": {
        "COGNITIVE_BROWSER_MODEL": "cross-encoder/mmarco-mMiniLMv2-L12-H384-v1",
        "COGNITIVE_BROWSER_DEVICE": "cuda"
      }
    }
  }
}