webbrain

Features

• Page Reading — Extracts text, links, forms, tables, and interactive elements from any page
• Browser Actions — Click, type, scroll, navigate, and interact with page elements
• Ask / Act / Dev Modes — Read-only by default, normal browser actions on request, and Dev add-ons for source/style/page-debugging work
• Plan before Act — Act and Dev modes can generate a structured plan, show it for approval, then pin the approved plan to the scratchpad before tools run
• Multi-Step Agent — Autonomous task execution with tool-use loops (configurable, default 130 steps)
• Continue from Limit — When the agent hits the step limit, click Continue to keep going
• Multi-Provider LLM — Supports local and cloud models:
  • WebBrain Cloud 1.0 (cloud, default) — Built-in managed cloud option; no local setup required
  • llama.cpp (local) — No API key needed. Also Ollama, LM Studio, Jan, vLLM, SGLang, and LocalAI
  • OpenAI (GPT-5.5, etc.)
  • Anthropic Claude (native API)
  • Google Gemini, Mistral AI, DeepSeek, xAI Grok, Groq
  • MiniMax, Alibaba Cloud (Qwen)
  • Cloudflare Workers AI, Nvidia NIM
  • OpenRouter (default model: openrouter/free; access 100+ models)
• Onboarding Wizard — First-launch walkthrough covering Act mode safety and provider setup
• Side Panel UI — Clean chat interface that lives alongside your browsing
• Per-Tab Conversations — Each tab has its own chat history
• User Memory — Optional local memory for user-stated preferences, with explicit /memory --add commands and opt-in background auto-learning
• Streaming — Real-time token streaming from all providers
• Smart Context — Token-aware auto-compaction (summarizes older turns once the conversation nears the model's context window, with a visible "Context automatically compacted" notice), tool result limits, and emergency overflow recovery
• Browser History Control — Act mode can use native go_back / go_forward history tools instead of CSP-sensitive page JavaScript
• API Shortcut Hints — Repeated clicks that fire the same XHR/fetch request can surface a matching fetch_url suggestion while preserving the UI-first and /allow-api mutation policy
• On-demand Skills and Skill Tools — Settings → Skills can import trusted skill text or URLs. Mid/Full runs receive a small eligible ID/name/summary/semantic-intent catalog and load full instructions plus compatible webbrain-tools only when relevant; Compact disables skills. FreeSkillz.xyz and the browser-only email verification-code helper are enabled by default, and either can be removed.
• Copy Support — Copy buttons on code blocks and full messages
• Page Inspection Banner — Visual indicator when the agent is interacting with the page
• Stop Button — Abort the agent mid-execution at any time
• Deterministic Action Modes — Act and Dev modes use temperature 0.15 for browser-control decisions; Ask mode uses 0.3, and dedicated vision screenshot descriptions use 0

Quick Start

Chrome

git clone https://github.com/webbrain-one/webbrain.git

1. Open Chrome → chrome://extensions/
2. Enable Developer mode (top right)
3. Click Load unpacked → select the webbrain/src/chrome folder

Firefox

git clone https://github.com/webbrain-one/webbrain.git

1. Open Firefox → about:debugging#/runtime/this-firefox
2. Click Load Temporary Add-on
3. Navigate to src/firefox/ and select manifest.json

Note: Temporary add-ons are removed when Firefox restarts. For permanent installation, the extension needs to be signed via addons.mozilla.org.

Start a local LLM (default)

# Using llama.cpp
llama-server -m your-model.gguf --port 8080

# Or using Ollama (OpenAI-compatible)
ollama serve
# Then set base URL to http://localhost:11434/v1 in settings
# Or run: ollama launch webbrain --model <model>

# Or using Jan (OpenAI-compatible)
# Start Jan's local API server and use http://localhost:1337/v1

# Or using vLLM / SGLang (OpenAI-compatible)
vllm serve your-model --port 8000
python -m sglang.launch_server --model-path your-model --port 30000

Ollama launch handoff (preview)

WebBrain supports Ollama today through the local OpenAI-compatible provider. A new ollama launch webbrain --model <model> handoff can also configure WebBrain automatically, but it is not integrated into upstream Ollama yet. For now, try it from the codex/ollama-webbrain-launch-handoff branch of esokullu/ollama; we hope Ollama will integrate it upstream.

git clone https://github.com/esokullu/ollama.git
cd ollama
git switch codex/ollama-webbrain-launch-handoff
cmake -S . -B build -G Ninja -DOLLAMA_MLX_BACKENDS=
cmake --build build --parallel 8

OLLAMA_ORIGINS="chrome-extension://*,moz-extension://*" ./ollama serve
./ollama launch webbrain --model <model>

Context window: For reliable agent runs, load a local model with at least a 16k-token context window (the usable minimum). 8k can work with Compact mode enabled (Settings → per-provider Prompt tier); 4k is too small to hold the system prompt + tool schemas. WebBrain auto-compacts the conversation as it nears the window. Local providers default to 16k unless you set an explicit size in Settings. Test connection / Load models auto-detect the real window for llama.cpp, Ollama, and LM Studio when those backends report it (llama.cpp /props, Ollama /api/ps then /api/show num_ctx, LM Studio /api/v0/models). Detection refreshes the default 16k; it shrinks a larger manual override only from live/runtime context (llama.cpp /props, Ollama /api/ps, LM Studio loaded context). Other local backends (Jan, vLLM, SGLang, LocalAI) keep the manual/default value.

Use it

Click the WebBrain icon → the side panel opens. Type a message like:

• "Summarize this page"
• "Find all links about pricing"
• "Fill in the search box with 'AI agents' and click Search"
• "Navigate to github.com and find trending repositories"

Configuration

Click the gear icon or go to the extension's Options page to configure:

Display Settings:

• Verbose Mode — Show full tool call JSON (off by default)
• Auto-screenshot — Provide visual context when DOM/page reads are insufficient
• Max Agent Steps — Configurable step limit (5-200, default 60)
• Plan before Act — Optionally generate and review a structured Act-mode plan before browser tools run (try mode by default; explicit off is preserved)

Profile and Memory:

• Profile auto-fill and user memory are stored in plaintext browser local storage.
• User memory can be managed from Settings -> Profile or with /memory, /memory --add <text>, and /memory --forget <id>.
• When enabled, active memory records are sent to the configured LLM provider as part of the system prompt; optional auto-learning makes a best-effort provider call only after a turn completes.

Skills:

• FreeSkillz.xyz ships enabled by default and can expose read_youtube_transcript, fetch_nytimes_article, resolve_public_media, and download_public_media through its skill manifest. On NYTimes/The Athletic tabs it is preactivated for the current run so a structured blocking pageGate can route directly to the credentialless article fallback; remove it from Settings → Skills if you do not want it available.
• The OTP / verification-code helper also ships enabled by default and loads only for relevant requests. It declares no network tool: on the active run tab, it prefers selected text or a bounded accessibility-tree subtree, matches the newest relevant service code, excludes SMS/native-app access, and honors Strict secret handling. When used, the scoped page content and code are included in the normal request to your configured LLM provider. If Record traces is enabled, raw tool results and model responses are also stored locally until those traces are deleted. Remove the skill from Settings → Skills if you do not want this guidance available.
• Imported skills are copied into browser local storage. Mid/Full runs send eligible IDs, names, summaries, and optional canonical semantic intents to the planner and load_skill catalog; full instructions are appended to the system prompt only after acti