Open-source AI browser agent for Chrome and Firefox (monorepo) 🧠
JavaScript ai-agentai-agentsai-toolsbrowser-aibrowser-automationbrowser-extensionchrome-aichrome-devtoolschrome-extensionchrome-extensionsfirefox-addonfirefox-addonsfirefox-extensionfirefox-extensionsgemmagemma4minimaxoffline-aioffline-llmqwen
Docs: decision model modes, in-browser vision choice, header permission switch, proxy context
Claude61983ae04296
diff --git a/web/docs/apocalypse-mode/index.html b/web/docs/apocalypse-mode/index.htmlindex ea439cf54..24c066f01 100644--- a/web/docs/apocalypse-mode/index.html+++ b/web/docs/apocalypse-mode/index.html@@ -19,15 +19,15 @@<aside class="docs-sidebar" aria-label="Documentation"><div class="sidebar-group"><div class="sidebar-label">Start here</div><nav><a href="/docs/">Overview</a><a href="/docs/faq/">FAQ</a></nav></div><div class="sidebar-group"><div class="sidebar-label">Use WebBrain</div><nav><a href="/docs/settings/">Settings, explained</a><a href="/docs/providers/">Providers & models</a><a href="/docs/easy-cli-proxy/">Subscription proxy</a><a href="/docs/safety/">Modes, safety & privacy</a><a href="/docs/apocalypse-mode/" aria-current="page">Apocalypse Mode</a><a href="/docs/formats/">Exports & workflows</a><a href="/docs/mcp/">MCP server</a><a href="/docs/lm-studio/">LM Studio plugin</a><a href="/docs/ollama/">Ollama</a></nav></div><div class="sidebar-group"><div class="sidebar-label">For builders</div><nav><a href="https://github.com/webbrain-one/webbrain">Source code ↗</a></nav></div><div class="sidebar-group"><div class="sidebar-label">Community</div><nav><a href="https://discord.gg/cgC325ssfw">Discord Support</a></nav></div></aside><main class="doc-main" id="main"><div class="breadcrumb"><a href="/docs/">Docs</a> / Apocalypse Mode</div>- <section class="doc-hero"><div class="eyebrow">WebBrain, ready when the internet isn’t</div><h1>Prepare local intelligence and practical knowledge before you need them.</h1><p class="lede">Apocalypse Mode assembles an offline-ready kit inside your browser: on-device text and vision models on supported Chromium systems, a searchable Wikipedia archive, and an Emergency Box of books and field references.</p><div class="mode-strip" aria-label="The three essentials in the Chromium Apocalypse Mode setup"><div class="mode ask"><strong>Text model</strong><span>1.55 GB · local WebGPU inference</span></div><div class="mode act"><strong>Vision model</strong><span>about 810 MB · screenshots stay local</span></div><div class="mode dev"><strong>Wikipedia</strong><span>Simple English · searchable offline</span></div></div></section>+ <section class="doc-hero"><div class="eyebrow">WebBrain, ready when the internet isn’t</div><h1>Prepare local intelligence and practical knowledge before you need them.</h1><p class="lede">Apocalypse Mode assembles an offline-ready kit inside your browser: on-device text and vision models on supported Chromium systems, a searchable Wikipedia archive, and an Emergency Box of books and field references.</p><div class="mode-strip" aria-label="The three essentials in the Chromium Apocalypse Mode setup"><div class="mode ask"><strong>Text model</strong><span>1.55 GB · local WebGPU inference</span></div><div class="mode act"><strong>Vision model</strong><span>optional · screenshots stay local</span></div><div class="mode dev"><strong>Wikipedia</strong><span>Simple English · searchable offline</span></div></div></section><section class="doc-section" id="meaning"><h2>What “offline” means</h2><p>You need an internet connection for the first download. Once the model files, Wikipedia archive, and chosen documents are stored, their core inference, search, and reading paths run locally from the browser profile. WebGPU performs model inference on the device; downloaded screenshots do not need a remote vision service.</p><div class="callout tip"><strong class="callout-title">Prepare on an ordinary day</strong><p>Large downloads and storage checks are much easier while power, connectivity, and time are predictable. Enable the mode, let every required item reach <strong>Ready</strong>, and test the readers before treating the kit as available.</p></div><p>Apocalypse Mode does not make every website or WebBrain task work offline. Live pages, hosted model providers, current news, and any resource you did not download still require a network. It prepares a self-contained local core; it is not a copy of the internet.</p></section>- <section class="doc-section" id="browsers"><h2>The Chromium and Firefox builds are different</h2><div class="table-wrap"><table><thead><tr><th>Build</th><th>What enabling the mode prepares</th><th>Ready and Emergency Box</th></tr></thead><tbody><tr><td>Chromium, with compatible WebGPU hardware</td><td>Starts the fixed local text model, local vision model, and the newest Simple English text-only Wikipedia archive. Model transfers share one local worker and run in sequence.</td><td><strong>Ready</strong> requires all three essentials. The Emergency Box unlocks when they are complete.</td></tr><tr><td>Firefox</td><td>Provides the offline Wikipedia library, reader, and Emergency Box. The Chromium-only WebGPU text and vision download stack is not shown.</td><td>Archive and document controls remain available without pretending Firefox has the Chromium WebGPU runtime.</td></tr></tbody></table></div><p>If the text and vision cards do not appear, first confirm that you are using the Chromium build on hardware and a browser version with WebGPU support.</p></section>+ <section class="doc-section" id="browsers"><h2>The Chromium and Firefox builds are different</h2><div class="table-wrap"><table><thead><tr><th>Build</th><th>What enabling the mode prepares</th><th>Ready and Emergency Box</th></tr></thead><tbody><tr><td>Chromium, with compatible WebGPU hardware</td><td>Starts the local text model and the newest Simple English text-only Wikipedia archive. The vision model is optional and downloads only when you click its Download button. Model transfers share one local worker and run in sequence.</td><td><strong>Ready</strong> requires the text model and the offline library; the vision model is optional. The Emergency Box unlocks when they are complete.</td></tr><tr><td>Firefox</td><td>Provides the offline Wikipedia library, reader, and Emergency Box. The Chromium-only WebGPU text and vision download stack is not shown.</td><td>Archive and document controls remain available without pretending Firefox has the Chromium WebGPU runtime.</td></tr></tbody></table></div><p>If the text and vision cards do not appear, first confirm that you are using the Chromium build on hardware and a browser version with WebGPU support.</p></section>- <section class="doc-section" id="setup"><h2>Build the basic kit</h2><ol class="step-list"><li><strong>Open Apocalypse Mode.</strong> Use the nuclear-symbol Apocalypse Mode control in the Settings header.</li><li><strong>Check your storage and connection.</strong> The Chromium basics include a 1.55 GB text model, an approximately 810 MB vision model, and the current Simple English Wikipedia archive. Leave additional room for temporary download state and Emergency Box resources.</li><li><strong>Turn the mode on.</strong> On supported Chromium systems, WebBrain starts the text model first, queues vision behind it, and starts the required Wikipedia archive automatically.</li><li><strong>Keep the browser open for model downloads.</strong> You may switch tabs or close Settings, but Chrome must stay open while the local model transfer continues. Follow the per-item progress instead of relying only on the total storage number.</li><li><strong>Resolve anything not ready.</strong> Resume a paused item, retry an error, or free storage. Removing a completed model deletes its cached files and requires another download before offline use.</li><li><strong>Test the kit.</strong> Use the vision test, open the Wikipedia reader, search for an article, and open at least one installed PDF. A download is not a plan until you have proved you can read it.</li></ol></section>+ <section class="doc-section" id="setup"><h2>Build the basic kit</h2><ol class="step-list"><li><strong>Open Apocalypse Mode.</strong> Use the nuclear-symbol Apocalypse Mode control in the Settings header.</li><li><strong>Check your storage and connection.</strong> The Chromium basics include a 1.55 GB text model and the current Simple English Wikipedia archive. An in-browser vision model is optional: about 1.45 GB for Ternary Bonsai 2 2B Vision or about 810 MB for WebBrain VL 2 450M. Leave additional room for temporary download state and Emergency Box resources.</li><li><strong>Turn the mode on.</strong> On supported Chromium systems, WebBrain starts the text model and the required Wikipedia archive automatically. It never downloads a vision model on its own.</li><li><strong>Keep the browser open for model downloads.</strong> You may switch tabs or close Settings, but Chrome must stay open while the local model transfer continues. Follow the per-item progress instead of relying only on the total storage number.</li><li><strong>Resolve anything not ready.</strong> Resume a paused item, retry an error, or free storage. Removing a completed model deletes its cached files and requires another download before offline use.</li><li><strong>Test the kit.</strong> Use the vision test, open the Wikipedia reader, search for an article, and open at least one installed PDF. A download is not a plan until you have proved you can read it.</li></ol></section>- <section class="doc-section" id="models"><h2>What the local models do</h2><div class="card-grid"><div class="card"><span class="tab-key">Text · 1.55 GB</span><h3>LFM2.5 2.6B</h3><p>WebBrain downloads <code>LiquidAI/LFM2.5-2.6B-ONNX</code> in <code>q4f16</code> form for local text generation through WebGPU. Use the nuclear control in standalone chat after it is ready.</p></div><div class="card"><span class="tab-key">Vision · about 810 MB</span><h3>WebBrain VL 2 450M</h3><p>The local vision layer interprets screenshots on the device, turning visual evidence into context without sending the image to a remote vision provider.</p></div></div><div class="callout warning"><strong class="callout-title">Apocalypse Mode does not replace your selected provider</strong><p>The downloaded text model is a run-scoped option for standalone chat. Enabling or downloading it does not silently change the provider used by ordinary WebBrain conversations.</p></div></section>+ <section class="doc-section" id="models"><h2>What the local models do</h2><div class="card-grid"><div class="card"><span class="tab-key">Text · 1.55 GB</span><h3>LFM2.5 2.6B</h3><p>WebBrain downloads <code>LiquidAI/LFM2.5-2.6B-ONNX</code> in <code>q4f16</code> form for local text generation through WebGPU. Use the nuclear control in standalone chat after it is ready.</p></div><div class="card"><span class="tab-key">Vision · optional</span><h3>Ternary Bonsai 2 2B Vision or WebBrain VL 2 450M</h3><p>The local vision layer interprets screenshots on the device, turning visual evidence into context without sending the image to a remote vision provider. The Vision Model card shares one choice with <strong>Settings → Assistive Models → Vision</strong>: <strong>Auto</strong> (default) uses Ternary Bonsai 2 2B Vision (about 1.45 GB) once it is downloaded; you can also pick WebBrain VL 2 450M (about 810 MB, vision only, needs consent in Settings first) or Off. If your Text Model is Ternary Bonsai 2 2B Vision, it already reads images and needs no extra download.</p></div></div><div class="callout warning"><strong class="callout-title">Apocalypse Mode does not replace your selected provider</strong><p>The downloaded text model is a run-scoped option for standalone chat. Enabling or downloading it does not silently change the provider used by ordinary WebBrain conversations.</p></div></section><section class="doc-section" id="wikipedia"><h2>Keep Wikipedia you can actually search</h2><p>The basic setup downloads the newest compact, text-only Simple English archive. Open <strong>Manage archives</strong> when you want another language, a fuller edition, or an edition with images. WebBrain resolves the archive’s current version, date, and exact size before installation.</p><div class="table-wrap"><table><thead><tr><th>Choice</th><th>Trade-off</th><th>Good for</th></tr></thead><tbody><tr><td>Text only</td><td>Smaller and faster to download; article text without the full image payload.</td><td>Compact reference, local factual retrieval, and broad language coverage.</td></tr><tr><td>With images</td><td>Uses substantially more storage.</td><td>Articles where diagrams, maps, and photographs matter.</td></tr><tr><td>Import a <code>.zim</code> file</td><td>You manage the source file and may need to reauthorize access after a restart.</td><td>Existing Kiwix collections or archives moved between machines.</td></tr></tbody></table></div><p>The Wikipedia reader searches article titles and opens the matching article from local storage. Local standalone chat can retrieve compact, attributed passages from the installed archive. When you replace the required Simple English archive, WebBrain keeps the old copy until the replacement has downloaded and verified successfully.</p></section>diff --git a/web/docs/assets/screenshots/settings-assistive-models.jpg b/web/docs/assets/screenshots/settings-assistive-models.jpgindex 33707267b..6f7671da0 100644Binary files a/web/docs/assets/screenshots/settings-assistive-models.jpg and b/web/docs/assets/screenshots/settings-assistive-models.jpg differdiff --git a/web/docs/assets/screenshots/settings-decision-models.jpg b/web/docs/assets/screenshots/settings-decision-models.jpgnew file mode 100644index 000000000..c47889127Binary files /dev/null and b/web/docs/assets/screenshots/settings-decision-models.jpg differdiff --git a/web/docs/assets/screenshots/settings-jev.jpg b/web/docs/assets/screenshots/settings-jev.jpgdeleted file mode 100644index a517e824d..000000000Binary files a/web/docs/assets/screenshots/settings-jev.jpg and /dev/null differdiff --git a/web/docs/easy-cli-proxy/index.html b/web/docs/easy-cli-proxy/index.htmlindex 101d0d5dc..4177f937c 100644--- a/web/docs/easy-cli-proxy/index.html+++ b/web/docs/easy-cli-proxy/index.html@@ -48,7 +48,7 @@<section class="doc-section" id="flow"><h2>The boundary in one line</h2><div class="bridge-map" aria-label="Request path from WebBrain to the model provider"><div class="bridge-node"><strong>WebBrain</strong><span>browser extension</span></div><div class="bridge-arrow" aria-hidden="true">→</div><div class="bridge-node local"><strong>127.0.0.1:8317</strong><span>EasyCLIProxyAPI + CLIProxyAPI</span></div><div class="bridge-arrow" aria-hidden="true">→</div><div class="bridge-node"><strong>Provider account</strong><span>remote inference and account rules</span></div></div><p>The first connection stays on your computer. The second does not. EasyCLIProxyAPI keeps upstream OAuth material outside WebBrain; WebBrain stores only the local proxy client key you enter in Settings.</p></section>- <section class="doc-section" id="setup"><div class="eyebrow">Setup</div><h2>Connect the local endpoint</h2><ol class="step-list"><li><strong>Inspect and install upstream.</strong> Download EasyCLIProxyAPI from its GitHub release page. In <strong>Version Management</strong>, install the matching CLIProxyAPI core.</li><li><strong>Lock the proxy to loopback.</strong> Keep the listener on <code>127.0.0.1</code>, leave LAN access off, and never forward port <code>8317</code>. Replace any short or default client key with a strong random value; the exact control may appear under Advanced Settings or API Access depending on the release.</li><li><strong>Authorize only the account you intend to use.</strong> Open <strong>OAuth</strong>, choose Codex, Claude, Antigravity, Kimi, or xAI, and finish the provider page. Stop if the provider presents terms or permissions you do not accept.</li><li><strong>Start the core and copy the OpenAI-compatible URL.</strong> On Home, start the runtime and copy the OpenAI endpoint. The common default is <code>http://127.0.0.1:8317/v1</code>; trust the value displayed by your running version.</li><li><strong>Open WebBrain Settings → Providers.</strong> Search for <strong>Local OpenAI-compatible Proxy</strong>. Paste the server URL and the proxy client key—not the provider OAuth token.</li><li><strong>Load and select a model.</strong> Click <strong>Load Models</strong>, select an exposed model, set its real context window, and enable vision only if that exact model accepts images.</li><li><strong>Save, test, then activate.</strong> Run <strong>Test Connection</strong> and require a green success result before selecting the provider for chat. Start with Ask mode on non-sensitive content.</li></ol><div class="callout warning"><strong class="callout-title">Do not use <code>0.0.0.0</code>, a LAN IP, tunnel, or public reverse proxy</strong><p>A local client key is not a safe internet-facing authentication system. Any process that obtains the key can send requests through the signed-in provider account.</p></div></section>+ <section class="doc-section" id="setup"><div class="eyebrow">Setup</div><h2>Connect the local endpoint</h2><ol class="step-list"><li><strong>Inspect and install upstream.</strong> Download EasyCLIProxyAPI from its GitHub release page. In <strong>Version Management</strong>, install the matching CLIProxyAPI core.</li><li><strong>Lock the proxy to loopback.</strong> Keep the listener on <code>127.0.0.1</code>, leave LAN access off, and never forward port <code>8317</code>. Replace any short or default client key with a strong random value; the exact control may appear under Advanced Settings or API Access depending on the release.</li><li><strong>Authorize only the account you intend to use.</strong> Open <strong>OAuth</strong>, choose Codex, Claude, Antigravity, Kimi, or xAI, and finish the provider page. Stop if the provider presents terms or permissions you do not accept.</li><li><strong>Start the core and copy the OpenAI-compatible URL.</strong> On Home, start the runtime and copy the OpenAI endpoint. The common default is <code>http://127.0.0.1:8317/v1</code>; trust the value displayed by your running version.</li><li><strong>Open WebBrain Settings → Providers.</strong> Search for <strong>Local OpenAI-compatible Proxy</strong>. Paste the server URL and the proxy client key—not the provider OAuth token.</li><li><strong>Load and select a model.</strong> Click <strong>Load Models</strong> and select an exposed model. WebBrain fills in <strong>Context window</strong> from the proxy’s model list, a llama.cpp server behind it, or the name of a known hosted model (128k at most); otherwise it uses 64k. It keeps a number you typed unless the server reports a smaller window, so lower it by hand if the proxy fronts a small local model that reports nothing. Enable vision only if that exact model accepts images.</li><li><strong>Save, test, then activate.</strong> Run <strong>Test Connection</strong> and require a green success result before selecting the provider for chat. Start with Ask mode on non-sensitive content.</li></ol><div class="callout warning"><strong class="callout-title">Do not use <code>0.0.0.0</code>, a LAN IP, tunnel, or public reverse proxy</strong><p>A local client key is not a safe internet-facing authentication system. Any process that obtains the key can send requests through the signed-in provider account.</p></div></section><section class="doc-section" id="troubleshooting"><h2>Troubleshooting by symptom</h2><div class="table-wrap"><table><thead><tr><th>Symptom</th><th>Likely cause</th><th>What to check</th></tr></thead><tbody><tr><td>Connection refused</td><td>The CLIProxyAPI core is stopped or the port changed.</td><td>Start the runtime, refresh Home, and copy the displayed OpenAI URL again.</td></tr><tr><td>Unauthorized</td><td>The WebBrain key does not match the proxy client key.</td><td>Copy the local client key again. Never paste an upstream OAuth token into WebBrain.</td></tr><tr><td>No models loaded</td><td>OAuth did not finish, the account exposes no compatible model, or upstream support changed.</td><td>Check the OAuth/account status and EasyCLIProxyAPI logs, then review its current release notes.</td></tr><tr><td>Text works but images fail</td><td>The selected alias does not accept image input.</td><td>Turn vision off or choose a model the proxy reports as vision-capable.</td></tr><tr><td>Quota or account error</td><td>The upstream account, plan, region, or provider policy rejected the request.</td><td>Stop retrying, inspect the provider account directly, and use an official API key when available.</td></tr><tr><td>Unexpected models or behavior after an update</td><td>The desktop app, core, plugin, or provider contract changed.</td><td>Compare versions, inspect release notes, and re-run Load Models and Test Connection.</td></tr></tbody></table></div></section>diff --git a/web/docs/faq/index.html b/web/docs/faq/index.htmlindex d70032bcf..829bc4b81 100644--- a/web/docs/faq/index.html+++ b/web/docs/faq/index.html@@ -209,7 +209,7 @@"name": "It's asking me too many questions. Can I disable that?","acceptedAnswer": {"@type": "Answer",- "text": "Those approval prompts are for your safety. Even on sites you trust, malicious actors can mimic LLM instructions or inject content that tries to make the agent do things on your behalf that you would not approve. That is why WebBrain asks before consequential actions. If you still do not want to go through those prompts, type /dangerously-skip-permissions in the side panel, or open the side plugin, click the wheel icon in the header, go to the Permissions tab in settings.html, and switch off Ask before consequential actions."+ "text": "Those approval prompts are for your safety. Even on sites you trust, malicious actors can mimic LLM instructions or inject content that tries to make the agent do things on your behalf that you would not approve. That is why WebBrain asks before consequential actions. If you still do not want to go through those prompts, type /dangerously-skip-permissions in the side panel, or open the side plugin, click the wheel icon in the header, go to the Permissions tab in settings.html, and switch off Ask before consequential actions. The ⚠ button in the side panel header opens the same switch."}},{@@ -697,7 +697,7 @@</details><details class="faq-question" id="disable-approval-questions" data-faq-item><summary><span>It's asking me too many questions. Can I disable that?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>- <div class="faq-answer"><p>Those approval prompts are for your safety. Even on sites you trust, malicious actors can mimic LLM instructions or inject content that tries to make the agent do things on your behalf that you would not approve. That is why WebBrain asks before consequential actions. If you still do not want to go through those prompts, type <code>/dangerously-skip-permissions</code> in the side panel, or open the side plugin, click the wheel icon in the header, go to the <strong>Permissions</strong> tab in <code>settings.html</code>, and switch off <strong>Ask before consequential actions</strong>.</p></div>+ <div class="faq-answer"><p>Those approval prompts are for your safety. Even on sites you trust, malicious actors can mimic LLM instructions or inject content that tries to make the agent do things on your behalf that you would not approve. That is why WebBrain asks before consequential actions. If you still do not want to go through those prompts, type <code>/dangerously-skip-permissions</code> in the side panel, or open the side plugin, click the wheel icon in the header, go to the <strong>Permissions</strong> tab in <code>settings.html</code>, and switch off <strong>Ask before consequential actions</strong>. The ⚠ button in the side panel header opens the same switch.</p></div></details><details class="faq-question" id="api-mutations" data-faq-item><summary><span>Does WebBrain call APIs directly, or does it always click through the UI?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>diff --git a/web/docs/index.html b/web/docs/index.htmlindex c897f77d3..8edb43002 100644--- a/web/docs/index.html+++ b/web/docs/index.html@@ -64,7 +64,7 @@<div class="walkthrough-copy"><div class="eyebrow">Watch first · 12 min 40 sec</div><h2 id="settings-walkthrough-title">Settings, from the useful defaults to the advanced controls.</h2>- <p>Start with the few choices most people need, then follow the Settings tabs through providers, supporting models, memory, skills, permissions, and the advanced drawer. The older video calls Assistive Models “Multimodal”; the current tab also contains Jev (TypeSafe).</p>+ <p>Start with the few choices most people need, then follow the Settings tabs through providers, supporting models, memory, skills, permissions, and the advanced drawer. The older video calls Assistive Models “Multimodal”; the current tab also contains Decision models.</p></div><div class="walkthrough-player"><video controls preload="metadata" playsinline poster="/assets/webbrain-settings-walkthrough-poster.jpg">@@ -120,7 +120,7 @@<div class="settings-map" aria-label="Settings depth map"><div class="setting-node everyday"><strong>General</strong><span>behavior & comfort</span></div><div class="setting-node everyday"><strong>Providers</strong><span>the active model</span></div>- <div class="setting-node sometimes"><strong>Assistive Models</strong><span>vision, speech & Jev</span></div>+ <div class="setting-node sometimes"><strong>Assistive Models</strong><span>vision, speech & decisions</span></div><div class="setting-node sometimes"><strong>Memory</strong><span>what persists</span></div><div class="setting-node expert"><strong>Skills</strong><span>trusted add-ons</span></div><div class="setting-node everyday"><strong>Permissions</strong><span>where it may act</span></div>diff --git a/web/docs/providers/index.html b/web/docs/providers/index.htmlindex 36512762c..77e7ac8de 100644--- a/web/docs/providers/index.html+++ b/web/docs/providers/index.html@@ -35,7 +35,7 @@<section class="doc-section" id="troubleshoot"><h2>When Test Connection fails</h2><ol><li><strong>Connection refused:</strong> start the local server and confirm the port.</li><li><strong>CORS or origin error:</strong> allow the WebBrain extension origin in the local server.</li><li><strong>Unauthorized:</strong> paste a current key for the same provider as the base URL.</li><li><strong>Model not found:</strong> copy the exact model ID from the provider or use Load models.</li><li><strong>Requests time out:</strong> increase the General request timeout for large local models on CPU.</li><li><strong>Runs fail after a few turns:</strong> confirm the real context window; lower the prompt tier or load the model with a larger context.</li></ol><div class="callout tip"><strong class="callout-title">Test before you switch</strong><p>Keep the working provider active while you configure a new one. Save and test the new card first; only then click Set Active.</p></div></section>- <section class="doc-section" id="assistive-models"><h2>Optional assistive models</h2><p>The active provider handles the main conversation, planning and final replies. Use <strong>Settings → Assistive Models</strong> for a separate <strong>Vision</strong> model, <strong>Speech to text</strong> endpoint or <strong>Jev (TypeSafe)</strong>. These settings are independent of provider filtering and selection.</p><p>Jev offers optional scheduled verification and separate default-off fast classifications and experimental browser actions. Existing permissions and completion checks still apply; live speed gains have not been established. See <a href="/docs/settings/#multimodal">Assistive Models settings</a> for setup and data-sharing details.</p></section>+ <section class="doc-section" id="assistive-models"><h2>Optional assistive models</h2><p>The active provider handles the main conversation, planning and final replies. Use <strong>Settings → Assistive Models</strong> for a separate <strong>Vision</strong> model, <strong>Speech to text</strong> endpoint or <strong>Decision models</strong>. These settings are independent of provider filtering and selection.</p><p>Vision and Decision models default to <strong>Auto</strong>: in Chrome they use Ternary Bonsai 2 2B Vision once it is downloaded and nothing before that. A decision model only answers yes-or-no and multiple-choice questions such as “Is this task finished?”; turn it <strong>On</strong> to use OpenRouter, TypeSafe, a local server, D1 or the in-browser model for every decision, or <strong>Off</strong> to never use one. Existing permissions and completion checks still apply. See <a href="/docs/settings/#multimodal">Assistive Models settings</a> for setup and data-sharing details.</p></section><nav class="next-links" aria-label="Next guides"><a href="/docs/settings/"><span>Back</span>← Settings, explained</a><a href="/docs/safety/"><span>Next</span>Modes, safety & privacy →</a></nav></main><aside class="on-this-page" aria-label="On this page"><h2>On this page</h2><a href="#quick-choice">Quick choice</a><a href="#catalog">Provider catalog</a><a href="#setup">Configure</a><a href="#fields">Fields</a><a href="#local">Local models</a><a href="#tiers">Prompt tiers</a><a href="#troubleshoot">Troubleshoot</a></aside>diff --git a/web/docs/settings/index.html b/web/docs/settings/index.htmlindex 9c7f5698b..293e8416c 100644--- a/web/docs/settings/index.html+++ b/web/docs/settings/index.html@@ -24,7 +24,7 @@<div class="breadcrumb"><a href="/docs/">Docs</a> / Settings</div><section class="doc-hero"><div class="eyebrow">Settings, without guesswork</div><h1>Know what to change—and what to leave alone.</h1><p class="lede">Open Settings from the gear in the side panel or from your browser’s extension options. Most people only need General, Providers, and Permissions; the other tabs solve specific problems.</p><div class="meta">Screenshots from the Assistive Models update · synthetic configuration, no real credentials</div></section>- <section class="doc-section" id="defaults"><h2>A good default setup</h2><div class="table-wrap"><table><thead><tr><th>Control</th><th>Recommended</th><th>Why</th></tr></thead><tbody><tr><td>Plan before Act</td><td>Try planning</td><td>Uses a plan when available and safely falls back to a read-only turn if planner formatting fails.</td></tr><tr><td>Plan review stops</td><td>Below confidence threshold</td><td>Low-confidence plans pause; confident plans can continue.</td></tr><tr><td>Review below confidence</td><td>90%</td><td>A conservative balance between supervision and flow.</td></tr><tr><td>Max Agent Steps</td><td>130</td><td>Enough room for multi-step work. Stop the run yourself if behavior looks repetitive.</td></tr><tr><td>LLM request timeout</td><td>120 seconds</td><td>Cloud models usually answer sooner; large local models may need the extra startup time.</td></tr><tr><td>Scheduled action confirmation</td><td>On</td><td>Unattended work still pauses before consequential actions.</td></tr><tr><td>Permission gate</td><td>On</td><td>WebBrain asks before it first acts on a site and remembers your explicit choice.</td></tr></tbody></table></div></section>+ <section class="doc-section" id="defaults"><h2>A good default setup</h2><div class="table-wrap"><table><thead><tr><th>Control</th><th>Recommended</th><th>Why</th></tr></thead><tbody><tr><td>Plan before Act</td><td>Try planning</td><td>Uses a plan when available and safely falls back to a read-only turn if planner formatting fails.</td></tr><tr><td>Plan review stops</td><td>Below confidence threshold</td><td>Low-confidence plans pause; confident plans can continue.</td></tr><tr><td>Review below confidence</td><td>90%</td><td>A conservative balance between supervision and flow.</td></tr><tr><td>Max Agent Steps</td><td>130</td><td>Enough room for multi-step work. Stop the run yourself if behavior looks repetitive.</td></tr><tr><td>LLM request timeout</td><td>120 seconds</td><td>Cloud models usually answer sooner; large local models may need the extra startup time.</td></tr><tr><td>Scheduled action confirmation</td><td>On</td><td>Unattended work still pauses before consequential actions.</td></tr><tr><td>Permission gate</td><td>On</td><td>WebBrain asks before it first acts on a site and remembers your explicit choice. The ⚠ button in the side panel header switches it too.</td></tr><tr><td>Decision models</td><td>Auto</td><td>Nothing to set up. Uses Ternary Bonsai 2 2B Vision for completion checks once it is downloaded, and no decision model before that.</td></tr><tr><td>In-browser vision model</td><td>Auto</td><td>Reads screenshots with Ternary Bonsai 2 2B Vision once it is downloaded, only when your chat model cannot see images.</td></tr></tbody></table></div></section><section class="doc-section" id="general"><div class="eyebrow">Everyday</div><h2>General: how the agent behaves</h2><p>Language and appearance are cosmetic. The controls below them change the agent loop: how it plans, when you review, how long it can work, and whether scheduled jobs can run.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-general.jpg"><img src="/docs/assets/screenshots/settings-general.jpg" alt="WebBrain General settings showing language, appearance, planning, plan review and confidence controls"></a><figcaption>General starts with the controls you are most likely to use. Click the image to view it at full size.</figcaption></figure><h3>Planning controls</h3><ul><li><strong>Try planning</strong> is the best general choice. WebBrain attempts a structured plan before Act or Dev tools run; if planner formatting still fails after repair, that turn continues with the Ask prompt and read-only tools.</li><li><strong>Strict planning</strong> cancels the run when a valid plan cannot be produced. Use it where “no plan, no action” matters more than convenience.</li><li><strong>Off</strong> removes the extra full-plan call, while retaining the compact intent check and its safe read-only fallback. It can be faster and cheaper, but you lose the plan review checkpoint.</li></ul>@@ -39,17 +39,17 @@<section class="doc-section" id="captcha"><h2>CAPTCHA solvers</h2><p>General → Advanced has a dedicated block for CapSolver, 2Captcha, CapMonster Cloud, SolveCaptcha, Anti-Captcha, NopeCHA and NoneCap. Entering a key checks its enabled box; you can turn it off while keeping the key. Save persists that choice. Testing a balance does not enable solving.</p><p>Expand a provider’s Advanced section to adjust its weight. Defaults are 100 through 94 in the order above. Enable any number; compatible providers run from highest weight to lowest. Errors and timeouts can cause more than one service to charge.</p><p>hCaptcha uses NopeCHA or NoneCap. Other families have provider-specific coverage and input requirements. See the <a href="/docs/captcha/">CAPTCHA guide</a> for the full coverage table, fallback behavior, privacy and page-application limits.</p></section><section class="doc-section" id="providers"><div class="eyebrow">Everyday</div><h2>Providers: choose the brain</h2><p>Only one provider is active at a time. Filter the list by Local, Cloud, or Router; search by provider, model, or field; then open a card to edit it. The safe sequence is <strong>Save → Test Connection → Set Active</strong>.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-providers.jpg"><img src="/docs/assets/screenshots/settings-providers.jpg" alt="WebBrain Providers settings with category filters, search and the active provider card"></a><figcaption>The active provider remains visible even when category filters are applied. Example model names are illustrative.</figcaption></figure><p>See <a href="/docs/providers/">Providers & models</a> for local server URLs, context windows, prompt tiers, and troubleshooting.</p></section>- <section class="doc-section" id="multimodal"><div class="eyebrow">When needed</div><h2>Assistive Models: support the main agent</h2><p>Choose the main conversation model in <strong>Providers</strong>. Open <strong>Settings → Assistive Models</strong> for the optional models below; configuring one does not change your active provider.</p>- <h3>Vision</h3><p>Use a dedicated OpenAI-compatible endpoint for screenshots while the active provider handles planning and tools. Chrome also offers an optional local vision fallback. Leave both unset to use the active provider for vision. Screenshot limits and redaction are grouped here too.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-assistive-models.jpg"><img src="/docs/assets/screenshots/settings-assistive-models.jpg" alt="Assistive Models settings with Vision and its screenshot controls"></a><figcaption>Vision comes first, followed by Speech to text and Jev (TypeSafe). Optional local vision is available in Chrome.</figcaption></figure>+ <section class="doc-section" id="multimodal"><div class="eyebrow">When needed</div><h2>Assistive Models: support the main agent</h2><p>Choose the main conversation model in <strong>Providers</strong>. Open <strong>Settings → Assistive Models</strong> for the optional models below; configuring one does not change your active provider.</p><p>Vision and Decision models both default to <strong>Auto</strong>. In Chrome with hardware WebGPU, Auto uses <strong>Ternary Bonsai 2 2B Vision</strong> (about 1.45 GB, runs on your device) once it is downloaded, and nothing before that. The <strong>Download a local model</strong> button on the install page, <strong>Providers → WebGPU</strong>, the Vision card and Apocalypse Mode all download that same model, so one download covers chat, screenshots and decisions.</p>+ <h3>Vision</h3><p>Pick what reads screenshots. An OpenAI-compatible endpoint, if you save one, reads every screenshot while the active provider keeps planning and tool calls. Without one, the active model reads screenshots itself if it can see images; if it cannot, the in-browser vision model does (Chrome only). Leave everything at its default to let the active provider handle vision. Screenshot limits and redaction are grouped here too.</p><p>The in-browser vision model has four settings:</p><ul><li><strong>Auto</strong> (default): uses Ternary Bonsai 2 2B Vision once it is downloaded and stays off until then.</li><li><strong>Ternary Bonsai 2 2B Vision</strong>: about 1.45 GB. The same model also chats and answers decisions.</li><li><strong>WebBrain VL 2 450M</strong>: about 810 MB, vision only. Its first download asks for your consent and checks WebGPU.</li><li><strong>Off</strong>: no in-browser vision model.</li></ul><p>Changing the setting never downloads anything; use the card’s <strong>Download</strong> button. Tasks never wait for a download, and screenshots read in the browser stay on your device. Firefox has no in-browser vision model.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-assistive-models.jpg"><img src="/docs/assets/screenshots/settings-assistive-models.jpg" alt="Assistive Models settings: the Vision card with the in-browser vision model set to Auto and the optional OpenAI-compatible endpoint"></a><figcaption>A fresh install: the in-browser vision model is on Auto and stays off until Ternary Bonsai 2 2B Vision is downloaded.</figcaption></figure><h3>Speech to text</h3><p>Configure a Whisper-compatible endpoint for recordings. Leave it blank and WebBrain tries configured providers in this order: OpenAI, Groq, LM Studio, then llama.cpp.</p>- <h3>Jev (TypeSafe)</h3><p>Open <strong>Settings → Assistive Models → Jev (TypeSafe)</strong> to add a TypeSafe API key and enable the features you want. Scheduled-watch and scheduled-completion verification use limited page observations to check a result. Probability thresholds live in the card’s Advanced section.</p><p><strong>Fast classifications</strong> and <strong>Fast browser actions — Experimental</strong> are separate switches, off by default. They do not replace the main planner, permissions or completion checks. No general speed improvement has been established by live benchmarks.</p><p>Use <strong>Save</strong> to store changes, <strong>Test Connection</strong> for one request with fixed example data, or <strong>Delete key</strong> to remove the key and reset Jev options. Turning off Enable Jev preserves your individual preferences. Enabled features send limited task and page data to TypeSafe and may incur charges; Strict Secret Mode disables Jev.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-jev.jpg"><img src="/docs/assets/screenshots/settings-jev.jpg" alt="Jev settings with scheduled verification and separate optional fast-decision switches, all disabled"></a><figcaption>Jev is an optional assistive model. This example contains no API key; every Jev feature is disabled.</figcaption></figure>+ <h3>Decision models</h3><p>A decision model is a small, fast model that answers the agent’s yes-or-no and multiple-choice questions, such as “Is this task finished?”, so your chat model doesn’t have to. It never plans, clicks or types. Choose one mode:</p><ul><li><strong>Auto</strong> (default): nothing to set up. In Chrome it uses Ternary Bonsai 2 2B Vision once it is downloaded, to check whether side panel tasks and scheduled watches and tasks really finished. While WebBrain Cloud (Compass) is your chat provider, Compass’s decision model answers instead. Otherwise no decision model is used. The card shows which one is in use.</li><li><strong>On</strong>: uses the provider you set up for every decision: OpenRouter, TypeSafe, a local System One server, D1, or the in-browser chat model in Chrome. All decisions are on by default. Open <strong>Choose which decisions use it</strong> to turn off individual ones (Verify completion, Verify scheduled watches, Verify scheduled task completion, Fast classifications, Fast browser actions — Experimental) and to set thresholds.</li><li><strong>Off</strong>: never uses a decision model, Compass’s included. Your chat model still checks whether tasks are finished.</li></ul><p>A mode change applies at once; On’s provider fields use <strong>Save</strong>. <strong>Test Connection</strong> sends one request with fixed example data. <strong>Reset</strong> returns to Auto and deletes saved decision keys. Cloud decision providers receive limited task and page data and may cost money; Strict Secret Mode disables decision models. Profiles that had decision outsourcing turned on before these modes keep working as On with their previous switches.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-decision-models.jpg"><img src="/docs/assets/screenshots/settings-decision-models.jpg" alt="Decision models settings with Auto selected and the status line showing that no decision model is in use yet"></a><figcaption>Auto on a fresh install: nothing is downloaded yet, so no decision model is in use. Provider fields appear only with On.</figcaption></figure></section><section class="doc-section" id="memory"><div class="eyebrow">When needed</div><h2>Memory: decide what persists</h2><figure class="shot"><a href="/docs/assets/screenshots/settings-memory.jpg"><img src="/docs/assets/screenshots/settings-memory.jpg" alt="WebBrain Memory settings showing encrypted cloud sync and profile auto-fill"></a><figcaption>Memory groups three different features: encrypted sync, profile auto-fill, and user memory.</figcaption></figure><ul><li><strong>Encrypted Cloud Sync</strong> is for WebBrain Compass members. Supported provider keys, profile text, and memories are encrypted with your sync password before upload. WebBrain cannot recover that password. Chat history and OAuth sign-ins are not synced.</li><li><strong>Profile auto-fill</strong> is a reusable block of personal details. It is stored in plaintext in this browser profile and sent to the active provider when enabled.</li><li><strong>User memory</strong> stores durable preferences. Explicit <code>/memory --add</code> saves immediately. Auto-learn and learning from completed forms are optional.</li></ul><div class="callout danger"><strong class="callout-title">Do not turn memory into a password manager</strong><p>Do not store passwords, API keys, tokens, recovery codes, or other secrets in profile text or user memory.</p></div></section><section class="doc-section" id="skills"><div class="eyebrow">Trust decision</div><h2>Skills: extra instructions and tools</h2><p>A skill can add instructions to the system prompt. A skill containing a <code>webbrain-tools</code> manifest can also expose HTTP or download-job tools. Imported skill text is copied into browser local storage and sent to your configured LLM as part of the system prompt.</p><div class="callout warning"><strong class="callout-title">Importing is approval</strong><p>Only import a skill you trust. Its declared HTTPS tools can send their declared inputs to the declared endpoint without a new prompt for every call. Download tools still use the normal Downloads permission gate before saving a file.</p></div></section>- <section class="doc-section" id="permissions"><div class="eyebrow">Everyday safety</div><h2>Permissions: where WebBrain may act</h2><p>Keep <strong>Ask before consequential actions</strong> on. When you choose “Always allow” in the side panel, Settings records a specific capability for a specific site. A click grant is not automatically a typing, download, upload, JavaScript, or scheduling grant.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-permissions.jpg"><img src="/docs/assets/screenshots/settings-permissions.jpg" alt="WebBrain Permissions settings with the master safety switch and per-site capability grants"></a><figcaption>Revoke one capability to make WebBrain ask again next time, or clear the whole list.</figcaption></figure><div class="callout danger"><strong class="callout-title">The master switch is global</strong><p>Turning it off allows prompt-free actions on every site until you turn it back on. This is much broader than granting one capability to one host.</p></div></section>+ <section class="doc-section" id="permissions"><div class="eyebrow">Everyday safety</div><h2>Permissions: where WebBrain may act</h2><p>Keep <strong>Ask before consequential actions</strong> on. When you choose “Always allow” in the side panel, Settings records a specific capability for a specific site. A click grant is not automatically a typing, download, upload, JavaScript, or scheduling grant.</p><p>The same switch sits in the side panel: the <strong>⚠</strong> button in the header opens it, with a link to the site list. While prompts are off, that button stays orange with a dot. Switching prompts off there never approves a permission card that is already waiting; answer that card yourself.</p><figure class="shot"><a href="/docs/assets/screenshots/settings-permissions.jpg"><img src="/docs/assets/screenshots/settings-permissions.jpg" alt="WebBrain Permissions settings with the master safety switch and per-site capability grants"></a><figcaption>Revoke one capability to make WebBrain ask again next time, or clear the whole list.</figcaption></figure><div class="callout danger"><strong class="callout-title">The master switch is global</strong><p>Turning it off allows prompt-free actions on every site until you turn it back on. This is much broader than granting one capability to one host.</p></div></section><nav class="next-links" aria-label="Next guides"><a href="/docs/providers/"><span>Next</span>Providers & models →</a><a href="/docs/safety/"><span>Also important</span>Modes, safety & privacy →</a></nav></main>diff --git a/web/docs/tr/faq/index.html b/web/docs/tr/faq/index.htmlindex 1ceb67d33..ba1a9ba91 100644--- a/web/docs/tr/faq/index.html+++ b/web/docs/tr/faq/index.html@@ -209,7 +209,7 @@"name": "Bana çok fazla soru soruyor. Bunu devre dışı bırakabilir miyim?","acceptedAnswer": {"@type": "Answer",- "text": "Bu onay istemleri güvenliğiniz içindir. Güvendiğiniz sitelerde bile kötü niyetli kişiler LLM talimatlarını taklit edebilir veya ajanın sizin adınıza onaylamayacağınız şeyler yapmasını sağlamaya çalışan içerikler enjekte edebilir. Bu yüzden WebBrain sonuç doğuran eylemlerden önce onayınızı ister. Yine de bu istemlerden geçmek istemiyorsanız, yan panele /dangerously-skip-permissions yazın veya devre dışı bırakabilirsiniz: yan eklentiyi açın, başlıktaki tekerlek simgesine tıklayın, settings.html içindeki Permissions sekmesine gidin ve Ask before consequential actions seçeneğini kapatın."+ "text": "Bu onay istemleri güvenliğiniz içindir. Güvendiğiniz sitelerde bile kötü niyetli kişiler LLM talimatlarını taklit edebilir veya ajanın sizin adınıza onaylamayacağınız şeyler yapmasını sağlamaya çalışan içerikler enjekte edebilir. Bu yüzden WebBrain sonuç doğuran eylemlerden önce onayınızı ister. Yine de bu istemlerden geçmek istemiyorsanız, yan panele /dangerously-skip-permissions yazın veya devre dışı bırakabilirsiniz: yan eklentiyi açın, başlıktaki tekerlek simgesine tıklayın, settings.html içindeki Permissions sekmesine gidin ve Ask before consequential actions seçeneğini kapatın. Yan panel başlığındaki ⚠ düğmesi de aynı anahtarı açar."}},{@@ -697,7 +697,7 @@</details><details class="faq-question" id="disable-approval-questions" data-faq-item><summary><span>Bana çok fazla soru soruyor. Bunu devre dışı bırakabilir miyim?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>- <div class="faq-answer"><p>Bu onay istemleri güvenliğiniz içindir. Güvendiğiniz sitelerde bile kötü niyetli kişiler LLM talimatlarını taklit edebilir veya ajanın sizin adınıza onaylamayacağınız şeyler yapmasını sağlamaya çalışan içerikler enjekte edebilir. Bu yüzden WebBrain sonuç doğuran eylemlerden önce onayınızı ister. Yine de bu istemlerden geçmek istemiyorsanız, yan panele <code>/dangerously-skip-permissions</code> yazın veya devre dışı bırakabilirsiniz: yan eklentiyi açın, başlıktaki tekerlek simgesine tıklayın, <code>settings.html</code> içindeki <strong>Permissions</strong> sekmesine gidin ve <strong>Ask before consequential actions</strong> seçeneğini kapatın.</p></div>+ <div class="faq-answer"><p>Bu onay istemleri güvenliğiniz içindir. Güvendiğiniz sitelerde bile kötü niyetli kişiler LLM talimatlarını taklit edebilir veya ajanın sizin adınıza onaylamayacağınız şeyler yapmasını sağlamaya çalışan içerikler enjekte edebilir. Bu yüzden WebBrain sonuç doğuran eylemlerden önce onayınızı ister. Yine de bu istemlerden geçmek istemiyorsanız, yan panele <code>/dangerously-skip-permissions</code> yazın veya devre dışı bırakabilirsiniz: yan eklentiyi açın, başlıktaki tekerlek simgesine tıklayın, <code>settings.html</code> içindeki <strong>Permissions</strong> sekmesine gidin ve <strong>Ask before consequential actions</strong> seçeneğini kapatın. Yan panel başlığındaki ⚠ düğmesi de aynı anahtarı açar.</p></div></details><details class="faq-question" id="api-mutations" data-faq-item><summary><span>WebBrain API'leri doğrudan mı çağırır, yoksa her zaman arayüzden mi gider?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>diff --git a/web/docs/zh/apocalypse-mode/index.html b/web/docs/zh/apocalypse-mode/index.htmlindex 0a4b434b0..3fee3a7ce 100644--- a/web/docs/zh/apocalypse-mode/index.html+++ b/web/docs/zh/apocalypse-mode/index.html@@ -18,15 +18,15 @@<aside class="docs-sidebar" aria-label="文档目录"><div class="sidebar-group"><div class="sidebar-label">从这里开始</div><nav><a href="/docs/zh/">概览</a><a href="/docs/zh/faq/">常见问题</a></nav></div><div class="sidebar-group"><div class="sidebar-label">使用 WebBrain</div><nav><a href="/docs/zh/settings/">设置说明</a><a href="/docs/zh/providers/">提供商与模型</a><a href="/docs/zh/easy-cli-proxy/">订阅代理</a><a href="/docs/zh/safety/">模式、安全与隐私</a><a href="/docs/zh/apocalypse-mode/" aria-current="page">末日模式</a><a href="/docs/zh/formats/">导出与工作流</a></nav></div><div class="sidebar-group"><div class="sidebar-label">面向开发者</div><nav><a href="https://github.com/webbrain-one/webbrain">源代码 ↗</a></nav></div><div class="sidebar-group"><div class="sidebar-label">社区</div><nav><a href="https://discord.gg/cgC325ssfw">Discord 支持</a></nav></div></aside><main class="doc-main" id="main"><div class="breadcrumb"><a href="/docs/zh/">文档</a> / 末日模式</div>- <section class="doc-hero"><div class="eyebrow">WebBrain,当互联网不可用时也能使用</div><h1>提前准备好本地智能和实用知识。</h1><p class="lede">末日模式在你的浏览器中组装一套离线就绪的工具包:在支持的 Chromium 系统上运行设备端文本和视觉模型,可搜索的维基百科存档,以及包含书籍和现场参考资料的应急箱。</p><div class="mode-strip" aria-label="Chromium 末日模式设置中的三个核心要素"><div class="mode ask"><strong>文本模型</strong><span>1.55 GB · 本地 WebGPU 推理</span></div><div class="mode act"><strong>视觉模型</strong><span>约 810 MB · 截图保留在本地</span></div><div class="mode dev"><strong>维基百科</strong><span>简易英语版 · 可离线搜索</span></div></div></section>+ <section class="doc-hero"><div class="eyebrow">WebBrain,当互联网不可用时也能使用</div><h1>提前准备好本地智能和实用知识。</h1><p class="lede">末日模式在你的浏览器中组装一套离线就绪的工具包:在支持的 Chromium 系统上运行设备端文本和视觉模型,可搜索的维基百科存档,以及包含书籍和现场参考资料的应急箱。</p><div class="mode-strip" aria-label="Chromium 末日模式设置中的三个核心要素"><div class="mode ask"><strong>文本模型</strong><span>1.55 GB · 本地 WebGPU 推理</span></div><div class="mode act"><strong>视觉模型</strong><span>可选 · 截图保留在本地</span></div><div class="mode dev"><strong>维基百科</strong><span>简易英语版 · 可离线搜索</span></div></div></section><section class="doc-section" id="meaning"><h2>"离线"的含义</h2><p>首次下载需要互联网连接。模型文件、维基百科存档和所选文档存储后,其核心推理、搜索和阅读路径从浏览器配置文件本地运行。WebGPU 在设备上执行模型推理;下载的截图不需要远程视觉服务。</p><div class="callout tip"><strong class="callout-title">在平常的日子里准备</strong><p>在电力、网络和时间都可预测时进行大文件下载和存储检查要容易得多。启用模式,让每个必要项目达到<strong>就绪</strong>状态,并在将工具包视为可用之前测试阅读器。</p></div><p>末日模式不会让每个网站或 WebBrain 任务都能离线工作。实时页面、托管模型提供商、当前新闻以及你未下载的任何资源仍需要网络。它准备了一个自包含的本地核心;它不是互联网的副本。</p></section>- <section class="doc-section" id="browsers"><h2>Chromium 和 Firefox 版本有所不同</h2><div class="table-wrap"><table><thead><tr><th>版本</th><th>启用模式后准备的内容</th><th>就绪与应急箱</th></tr></thead><tbody><tr><td>Chromium,具有兼容的 WebGPU 硬件</td><td>启动固定的本地文本模型、本地视觉模型和最新的简易英语纯文本维基百科存档。模型传输共享一个本地工作者并按顺序运行。</td><td><strong>就绪</strong>要求三个核心要素全部完成。应急箱在它们完成后解锁。</td></tr><tr><td>Firefox</td><td>提供离线维基百科库、阅读器和应急箱。不显示仅限 Chromium 的 WebGPU 文本和视觉下载栈。</td><td>在不假装 Firefox 拥有 Chromium WebGPU 运行时的情况下,存档和文档控件仍然可用。</td></tr></tbody></table></div><p>如果文本和视觉卡片没有显示,请首先确认你使用的是具有 WebGPU 支持的硬件和浏览器版本上的 Chromium 版本。</p></section>+ <section class="doc-section" id="browsers"><h2>Chromium 和 Firefox 版本有所不同</h2><div class="table-wrap"><table><thead><tr><th>版本</th><th>启用模式后准备的内容</th><th>就绪与应急箱</th></tr></thead><tbody><tr><td>Chromium,具有兼容的 WebGPU 硬件</td><td>启动本地文本模型和最新的简易英语纯文本维基百科存档。视觉模型是可选的,只有点击它的下载按钮时才会下载。模型传输共享一个本地工作者并按顺序运行。</td><td><strong>就绪</strong>要求文本模型和离线资料库完成;视觉模型是可选的。应急箱在它们完成后解锁。</td></tr><tr><td>Firefox</td><td>提供离线维基百科库、阅读器和应急箱。不显示仅限 Chromium 的 WebGPU 文本和视觉下载栈。</td><td>在不假装 Firefox 拥有 Chromium WebGPU 运行时的情况下,存档和文档控件仍然可用。</td></tr></tbody></table></div><p>如果文本和视觉卡片没有显示,请首先确认你使用的是具有 WebGPU 支持的硬件和浏览器版本上的 Chromium 版本。</p></section>- <section class="doc-section" id="setup"><h2>构建基础工具包</h2><ol class="step-list"><li><strong>打开末日模式。</strong>使用设置标题中的核符号末日模式控件。</li><li><strong>检查存储和连接。</strong>Chromium 基础套件包括 1.55 GB 文本模型、约 810 MB 视觉模型和当前的简易英语维基百科存档。为临时下载状态和应急箱资源留出额外空间。</li><li><strong>开启模式。</strong>在支持的 Chromium 系统上,WebBrain 先启动文本模型,将视觉模型排队在其后,并自动启动所需的维基百科存档。</li><li><strong>保持浏览器打开以进行模型下载。</strong>你可以切换标签页或关闭设置,但在本地模型传输继续期间必须保持 Chrome 打开。跟踪每个项目的进度,不要仅依赖总存储数字。</li><li><strong>解决任何未就绪的项目。</strong>恢复暂停的项目、重试错误或释放存储空间。移除已完成的模型会删除其缓存文件,需要在离线使用前再次下载。</li><li><strong>测试工具包。</strong>使用视觉测试,打开维基百科阅读器,搜索一篇文章,并打开至少一个已安装的 PDF。下载不等于计划,除非你已证明可以阅读它。</li></ol></section>+ <section class="doc-section" id="setup"><h2>构建基础工具包</h2><ol class="step-list"><li><strong>打开末日模式。</strong>使用设置标题中的核符号末日模式控件。</li><li><strong>检查存储和连接。</strong>Chromium 基础套件包括 1.55 GB 文本模型和当前的简易英语维基百科存档。浏览器内视觉模型是可选的:Ternary Bonsai 2 2B Vision 约 1.45 GB,WebBrain VL 2 450M 约 810 MB。为临时下载状态和应急箱资源留出额外空间。</li><li><strong>开启模式。</strong>在支持的 Chromium 系统上,WebBrain 会自动启动文本模型和所需的维基百科存档,但从不自行下载视觉模型。</li><li><strong>保持浏览器打开以进行模型下载。</strong>你可以切换标签页或关闭设置,但在本地模型传输继续期间必须保持 Chrome 打开。跟踪每个项目的进度,不要仅依赖总存储数字。</li><li><strong>解决任何未就绪的项目。</strong>恢复暂停的项目、重试错误或释放存储空间。移除已完成的模型会删除其缓存文件,需要在离线使用前再次下载。</li><li><strong>测试工具包。</strong>使用视觉测试,打开维基百科阅读器,搜索一篇文章,并打开至少一个已安装的 PDF。下载不等于计划,除非你已证明可以阅读它。</li></ol></section>- <section class="doc-section" id="models"><h2>本地模型的功能</h2><div class="card-grid"><div class="card"><span class="tab-key">文本 · 1.55 GB</span><h3>LFM2.5 2.6B</h3><p>WebBrain 下载 <code>LiquidAI/LFM2.5-2.6B-ONNX</code> 的 <code>q4f16</code> 形式,用于通过 WebGPU 进行本地文本生成。就绪后在独立聊天中使用核控件。</p></div><div class="card"><span class="tab-key">视觉 · 约 810 MB</span><h3>WebBrain VL 2 450M</h3><p>本地视觉层在设备上解读截图,将视觉证据转化为上下文,无需将图像发送到远程视觉提供商。</p></div></div><div class="callout warning"><strong class="callout-title">末日模式不会替代你选择的提供商</strong><p>下载的文本模型是独立聊天的运行时选项。启用或下载它不会静默更改普通 WebBrain 对话使用的提供商。</p></div></section>+ <section class="doc-section" id="models"><h2>本地模型的功能</h2><div class="card-grid"><div class="card"><span class="tab-key">文本 · 1.55 GB</span><h3>LFM2.5 2.6B</h3><p>WebBrain 下载 <code>LiquidAI/LFM2.5-2.6B-ONNX</code> 的 <code>q4f16</code> 形式,用于通过 WebGPU 进行本地文本生成。就绪后在独立聊天中使用核控件。</p></div><div class="card"><span class="tab-key">视觉 · 可选</span><h3>Ternary Bonsai 2 2B Vision 或 WebBrain VL 2 450M</h3><p>本地视觉层在设备上解读截图,将视觉证据转化为上下文,无需将图像发送到远程视觉提供商。视觉模型卡片与<strong>设置 → 辅助模型 → 视觉</strong>共用同一个选项:<strong>自动</strong>(默认)会在 Ternary Bonsai 2 2B Vision(约 1.45 GB)下载完成后使用它;你也可以选择 WebBrain VL 2 450M(约 810 MB,仅用于视觉,需先在设置中同意)或关闭。如果文本模型是 Ternary Bonsai 2 2B Vision,它本身就能读取图像,无需额外下载。</p></div></div><div class="callout warning"><strong class="callout-title">末日模式不会替代你选择的提供商</strong><p>下载的文本模型是独立聊天的运行时选项。启用或下载它不会静默更改普通 WebBrain 对话使用的提供商。</p></div></section><section class="doc-section" id="wikipedia"><h2>保留你可以真正搜索的维基百科</h2><p>基础设置下载最新的紧凑型纯文本简易英语存档。当你需要其他语言、更完整版本或包含图像的版本时,打开<strong>管理存档</strong>。WebBrain 会在安装前解析存档的当前版本、日期和确切大小。</p><div class="table-wrap"><table><thead><tr><th>选项</th><th>权衡</th><th>适用场景</th></tr></thead><tbody><tr><td>纯文本</td><td>更小、下载更快;不含完整图像负载的文章文本。</td><td>紧凑参考、本地事实检索和广泛的语言覆盖。</td></tr><tr><td>包含图像</td><td>使用大量存储空间。</td><td>图表、地图和照片很重要的文章。</td></tr><tr><td>导入 <code>.zim</code> 文件</td><td>你管理源文件,重启后可能需要重新授权访问。</td><td>现有的 Kiwix 集合或在机器之间移动的存档。</td></tr></tbody></table></div><p>维基百科阅读器搜索文章标题并从本地存储打开匹配的文章。本地独立聊天可以从已安装的存档中检索紧凑的、有出处的段落。当你替换必需的简易英语存档时,WebBrain 会保留旧副本,直到替换下载并验证成功。</p></section>diff --git a/web/docs/zh/easy-cli-proxy/index.html b/web/docs/zh/easy-cli-proxy/index.htmlindex 29086618c..029bc965b 100644--- a/web/docs/zh/easy-cli-proxy/index.html+++ b/web/docs/zh/easy-cli-proxy/index.html@@ -48,7 +48,7 @@<section class="doc-section" id="flow"><h2>一句话说明请求边界</h2><div class="bridge-map" aria-label="从 WebBrain 到模型提供商的请求路径"><div class="bridge-node"><strong>WebBrain</strong><span>浏览器扩展</span></div><div class="bridge-arrow" aria-hidden="true">→</div><div class="bridge-node local"><strong>127.0.0.1:8317</strong><span>EasyCLIProxyAPI + CLIProxyAPI</span></div><div class="bridge-arrow" aria-hidden="true">→</div><div class="bridge-node"><strong>提供商账户</strong><span>远程推理与账户规则</span></div></div><p>第一段连接留在你的电脑上,第二段并非如此。EasyCLIProxyAPI 在 WebBrain 之外保存上游 OAuth 信息;WebBrain 只保存你在设置中输入的本地代理客户端密钥。</p></section>- <section class="doc-section" id="setup"><div class="eyebrow">配置</div><h2>连接本地端点</h2><ol class="step-list"><li><strong>检查并安装上游程序。</strong>从 GitHub 发行页面下载 EasyCLIProxyAPI。在<strong>版本管理</strong>中安装匹配的 CLIProxyAPI 核心。</li><li><strong>把代理限制在回环地址。</strong>监听地址保持为 <code>127.0.0.1</code>,关闭局域网访问,绝不要转发端口 <code>8317</code>。用高强度随机值替换简短或默认的客户端密钥;根据版本不同,相关控件可能位于高级设置或 API 访问中。</li><li><strong>只授权你确实要使用的账户。</strong>打开 <strong>OAuth</strong>,选择 Codex、Claude、Antigravity、Kimi 或 xAI,然后在提供商页面完成登录。如果提供商展示了你不接受的条款或权限,请立即停止。</li><li><strong>启动核心并复制 OpenAI 兼容 URL。</strong>在首页启动运行时并复制 OpenAI 端点。常见默认值是 <code>http://127.0.0.1:8317/v1</code>;应以当前运行版本显示的值为准。</li><li><strong>打开 WebBrain 设置 → 提供商。</strong>搜索<strong>本地 OpenAI 兼容代理</strong>。粘贴服务器 URL 和代理客户端密钥,而不是提供商的 OAuth 令牌。</li><li><strong>加载并选择模型。</strong>点击<strong>加载模型</strong>,选择代理暴露的模型,填写真实上下文窗口;只有该模型确实接受图像时才启用视觉。</li><li><strong>保存、测试,再启用。</strong>运行<strong>测试连接</strong>并确认出现绿色成功结果,然后再把它设为聊天提供商。先在非敏感内容上使用询问模式测试。</li></ol><div class="callout warning"><strong class="callout-title">不要使用 <code>0.0.0.0</code>、局域网 IP、隧道或公开反向代理</strong><p>本地客户端密钥不是安全的互联网公开认证系统。任何获得密钥的进程都可以通过已登录的提供商账户发送请求。</p></div></section>+ <section class="doc-section" id="setup"><div class="eyebrow">配置</div><h2>连接本地端点</h2><ol class="step-list"><li><strong>检查并安装上游程序。</strong>从 GitHub 发行页面下载 EasyCLIProxyAPI。在<strong>版本管理</strong>中安装匹配的 CLIProxyAPI 核心。</li><li><strong>把代理限制在回环地址。</strong>监听地址保持为 <code>127.0.0.1</code>,关闭局域网访问,绝不要转发端口 <code>8317</code>。用高强度随机值替换简短或默认的客户端密钥;根据版本不同,相关控件可能位于高级设置或 API 访问中。</li><li><strong>只授权你确实要使用的账户。</strong>打开 <strong>OAuth</strong>,选择 Codex、Claude、Antigravity、Kimi 或 xAI,然后在提供商页面完成登录。如果提供商展示了你不接受的条款或权限,请立即停止。</li><li><strong>启动核心并复制 OpenAI 兼容 URL。</strong>在首页启动运行时并复制 OpenAI 端点。常见默认值是 <code>http://127.0.0.1:8317/v1</code>;应以当前运行版本显示的值为准。</li><li><strong>打开 WebBrain 设置 → 提供商。</strong>搜索<strong>本地 OpenAI 兼容代理</strong>。粘贴服务器 URL 和代理客户端密钥,而不是提供商的 OAuth 令牌。</li><li><strong>加载并选择模型。</strong>点击<strong>加载模型</strong>并选择代理暴露的模型。WebBrain 会根据代理的模型列表、其后的 llama.cpp 服务器或已知托管模型的名称(最多 128k)自动填写<strong>上下文窗口</strong>;都无法得知时使用 64k。除非服务器报告了更小的窗口,否则不会覆盖你手动填写的数值;如果代理后面是不报告窗口的小型本地模型,请手动调低。只有该模型确实接受图像时才启用视觉。</li><li><strong>保存、测试,再启用。</strong>运行<strong>测试连接</strong>并确认出现绿色成功结果,然后再把它设为聊天提供商。先在非敏感内容上使用询问模式测试。</li></ol><div class="callout warning"><strong class="callout-title">不要使用 <code>0.0.0.0</code>、局域网 IP、隧道或公开反向代理</strong><p>本地客户端密钥不是安全的互联网公开认证系统。任何获得密钥的进程都可以通过已登录的提供商账户发送请求。</p></div></section><section class="doc-section" id="troubleshooting"><h2>按症状排查问题</h2><div class="table-wrap"><table><thead><tr><th>症状</th><th>可能原因</th><th>检查方法</th></tr></thead><tbody><tr><td>连接被拒绝</td><td>CLIProxyAPI 核心未启动或端口已改变。</td><td>启动运行时,刷新首页,然后重新复制显示的 OpenAI URL。</td></tr><tr><td>未授权</td><td>WebBrain 中的密钥与代理客户端密钥不一致。</td><td>重新复制本地客户端密钥。绝不要把上游 OAuth 令牌粘贴到 WebBrain。</td></tr><tr><td>没有加载到模型</td><td>OAuth 未完成、账户未暴露兼容模型,或上游支持已经变化。</td><td>检查 OAuth/账户状态和 EasyCLIProxyAPI 日志,然后查看当前发行说明。</td></tr><tr><td>文本可用但图像失败</td><td>所选模型别名不接受图像输入。</td><td>关闭视觉,或选择代理报告为支持视觉的模型。</td></tr><tr><td>配额或账户错误</td><td>上游账户、套餐、地区或提供商政策拒绝了请求。</td><td>停止重试,直接检查提供商账户,并在可用时改用官方 API 密钥。</td></tr><tr><td>更新后模型或行为异常</td><td>桌面应用、核心、插件或提供商协议发生变化。</td><td>比较版本、查看发行说明,然后重新运行“加载模型”和“测试连接”。</td></tr></tbody></table></div></section>diff --git a/web/docs/zh/faq/index.html b/web/docs/zh/faq/index.htmlindex 893aee4eb..79b87286a 100644--- a/web/docs/zh/faq/index.html+++ b/web/docs/zh/faq/index.html@@ -209,7 +209,7 @@"name": "它问我的问题太多了。可以关闭吗?","acceptedAnswer": {"@type": "Answer",- "text": "这些批准提示是为了保护你的安全。即使在你信任的网站上,恶意行为者也可能伪造 LLM 指令或注入内容,试图让代理代表你执行你不会同意的操作。因此,WebBrain 在执行有后果的操作之前始终需要你的批准。如果你仍然不想经过这些提示,可以在侧边面板输入 /dangerously-skip-permissions,或者关闭它们:打开侧边插件,点击标题栏中的齿轮图标,进入 settings.html 的 Permissions 标签页,然后关闭 Ask before consequential actions 选项。"+ "text": "这些批准提示是为了保护你的安全。即使在你信任的网站上,恶意行为者也可能伪造 LLM 指令或注入内容,试图让代理代表你执行你不会同意的操作。因此,WebBrain 在执行有后果的操作之前始终需要你的批准。如果你仍然不想经过这些提示,可以在侧边面板输入 /dangerously-skip-permissions,或者关闭它们:打开侧边插件,点击标题栏中的齿轮图标,进入 settings.html 的 Permissions 标签页,然后关闭 Ask before consequential actions 选项。侧边栏标题栏的 ⚠ 按钮也能打开同一个开关。"}},{@@ -697,7 +697,7 @@</details><details class="faq-question" id="disable-approval-questions" data-faq-item><summary><span>它问我的问题太多了。可以关闭吗?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>- <div class="faq-answer"><p>这些批准提示是为了保护你的安全。即使在你信任的网站上,恶意行为者也可能伪造 LLM 指令或注入内容,试图让代理代表你执行你不会同意的操作。因此,WebBrain 在执行有后果的操作之前始终需要你的批准。如果你仍然不想经过这些提示,可以在侧边面板输入 <code>/dangerously-skip-permissions</code>,或者关闭它们:打开侧边插件,点击标题栏中的齿轮图标,进入 <code>settings.html</code> 的 <strong>Permissions</strong> 标签页,然后关闭 <strong>Ask before consequential actions</strong> 选项。</p></div>+ <div class="faq-answer"><p>这些批准提示是为了保护你的安全。即使在你信任的网站上,恶意行为者也可能伪造 LLM 指令或注入内容,试图让代理代表你执行你不会同意的操作。因此,WebBrain 在执行有后果的操作之前始终需要你的批准。如果你仍然不想经过这些提示,可以在侧边面板输入 <code>/dangerously-skip-permissions</code>,或者关闭它们:打开侧边插件,点击标题栏中的齿轮图标,进入 <code>settings.html</code> 的 <strong>Permissions</strong> 标签页,然后关闭 <strong>Ask before consequential actions</strong> 选项。侧边栏标题栏的 ⚠ 按钮也能打开同一个开关。</p></div></details><details class="faq-question" id="api-mutations" data-faq-item><summary><span>WebBrain 是直接调用 API,还是始终通过界面点击?</span><span class="faq-question-mark" aria-hidden="true"></span></summary>diff --git a/web/docs/zh/index.html b/web/docs/zh/index.htmlindex 6648f75c8..be5b2ceea 100644--- a/web/docs/zh/index.html+++ b/web/docs/zh/index.html@@ -20,7 +20,7 @@<section class="doc-hero"><div class="eyebrow">WebBrain 使用手册</div><h1>浏览器代理有很多开关,但你不需要先学会所有开关。</h1><p class="lede">先打开一个网页并提出问题。这份指南先带你用最短路径得到结果,再在真正需要时解释更深层的设置。</p><div class="path-grid"><a class="path-card" href="#first-run"><span class="tab-key">第一次使用</span><h2>先完成一个任务</h2><p>打开侧边栏、选择合适模式,不改任何设置也能开始。</p><span class="arrow">开始使用 →</span></a><a class="path-card" href="/docs/zh/settings/"><span class="tab-key">控制代理</span><h2>看懂设置</h2><p>每个标签页会改变什么、哪些默认值适合大多数人、哪些可以先忽略。</p><span class="arrow">打开设置指南 →</span></a><a class="path-card" href="/docs/zh/providers/"><span class="tab-key">选择模型</span><h2>配置提供商</h2><p>使用 WebBrain Compass、云 API、模型路由或本地模型。</p><span class="arrow">选择模型 →</span></a><a class="path-card" href="/docs/zh/safety/"><span class="tab-key">保持控制</span><h2>安全使用执行模式</h2><p>了解询问、执行、开发模式,网站权限、记忆边界和高风险覆盖项。</p><span class="arrow">查看安全指南 →</span></a><a class="path-card" href="/docs/zh/formats/"><span class="tab-key">迁移内容</span><h2>放心导出</h2><p>区分对话、轨迹、设置与已保存工作流文件,并了解哪些内容敏感。</p><span class="arrow">了解导出格式 →</span></a></div></section><section class="doc-section" id="first-run"><div class="eyebrow">第一次运行</div><h2>前十分钟</h2><ol class="step-list"><li><strong>打开你想处理的网页。</strong>点击 WebBrain 图标,让侧边栏与网页并排显示。</li><li><strong>先留在“询问”模式。</strong>例如:“总结本页”“列出重要日期”“比较表格中的套餐”。询问模式可以阅读,但不会点击或输入。</li><li><strong>需要改动网页时再切到“执行”。</strong>例如:“在搜索框输入 wireless headphones 并搜索”。在新网站进行关键操作前,WebBrain 会请求权限。</li><li><strong>观察过程。</strong>随时使用停止按钮或 <kbd>Escape</kbd> 中止。遇到意外跳转、输入或重复动作,应立即停止。</li><li><strong>有明确需求时才打开设置。</strong>内置云提供商与推荐安全默认值足够完成第一次任务。</li></ol><div class="callout tip"><strong class="callout-title">好的提示词要写清结果</strong><p>“找到退款政策,并告诉我促销商品是否适用”比“看看这个网站”更容易验证。写明目标,也写明 WebBrain 不应越过的边界。</p></div></section><section class="doc-section" id="modes"><h2>先选模式,再发提示词</h2><div class="mode-strip"><div class="mode ask"><strong>询问 Ask</strong><span>阅读、解释、比较、提取;不改变网页。</span></div><div class="mode act"><strong>执行 Act</strong><span>点击、输入、导航、下载、安排任务。</span></div><div class="mode dev"><strong>开发 Dev</strong><span>增加源码、样式、DOM、框架和调试工具。</span></div></div></section>- <section class="doc-section" id="settings-map"><div class="eyebrow">它是一张地图,不是检查清单</div><h2>六个设置标签页</h2><div class="settings-map" aria-label="设置深度地图"><div class="setting-node everyday"><strong>常规</strong><span>行为与体验</span></div><div class="setting-node everyday"><strong>提供商</strong><span>当前模型</span></div><div class="setting-node sometimes"><strong>辅助模型</strong><span>视觉、语音与 Jev</span></div><div class="setting-node sometimes"><strong>记忆</strong><span>持久信息</span></div><div class="setting-node expert"><strong>技能</strong><span>受信任扩展</span></div><div class="setting-node everyday"><strong>权限</strong><span>允许操作的网站</span></div></div><div class="depth-legend"><span>日常控制</span><span class="sometimes">按需使用</span><span class="expert">高级信任决定</span></div><div class="callout"><strong class="callout-title">安全基线</strong><p>“执行前计划”保留为“尝试计划”,计划审查使用置信度阈值,计划任务确认保持开启,“关键操作前询问”保持开启。其他选项可以先不改。</p></div></section>+ <section class="doc-section" id="settings-map"><div class="eyebrow">它是一张地图,不是检查清单</div><h2>六个设置标签页</h2><div class="settings-map" aria-label="设置深度地图"><div class="setting-node everyday"><strong>常规</strong><span>行为与体验</span></div><div class="setting-node everyday"><strong>提供商</strong><span>当前模型</span></div><div class="setting-node sometimes"><strong>辅助模型</strong><span>视觉、语音与决策</span></div><div class="setting-node sometimes"><strong>记忆</strong><span>持久信息</span></div><div class="setting-node expert"><strong>技能</strong><span>受信任扩展</span></div><div class="setting-node everyday"><strong>权限</strong><span>允许操作的网站</span></div></div><div class="depth-legend"><span>日常控制</span><span class="sometimes">按需使用</span><span class="expert">高级信任决定</span></div><div class="callout"><strong class="callout-title">安全基线</strong><p>“执行前计划”保留为“尝试计划”,计划审查使用置信度阈值,计划任务确认保持开启,“关键操作前询问”保持开启。其他选项可以先不改。</p></div></section><nav class="next-links" aria-label="后续指南"><a href="/docs/zh/settings/"><span>下一篇</span>设置说明 →</a><a href="/docs/zh/safety/"><span>推荐阅读</span>模式、安全与隐私 →</a></nav></main><aside class="on-this-page" aria-label="本页内容"><h2>本页内容</h2><a href="#first-run">前十分钟</a><a href="#modes">选择模式</a><a href="#settings-map">设置地图</a></aside>diff --git a/web/docs/zh/providers/index.html b/web/docs/zh/providers/index.htmlindex 5fad88f81..df5073993 100644--- a/web/docs/zh/providers/index.html+++ b/web/docs/zh/providers/index.html@@ -20,7 +20,7 @@<section class="doc-section" id="local"><h2>本地模型的额外环节</h2><div class="table-wrap"><table><thead><tr><th>服务器</th><th>常见 Base URL</th><th>API 密钥</th></tr></thead><tbody><tr><td>llama.cpp</td><td><code>http://localhost:8080</code></td><td>通常不需要</td></tr><tr><td>Ollama</td><td><code>http://localhost:11434/v1</code></td><td>不需要</td></tr><tr><td>ODS</td><td><code>http://localhost:11434/v1</code></td><td>可选;使用 ODS .env 中的端口</td></tr><tr><td>LM Studio</td><td><code>http://localhost:1234/v1</code></td><td>不需要</td></tr><tr><td>Jan</td><td><code>http://localhost:1337/v1</code></td><td>可选</td></tr><tr><td>vLLM</td><td><code>http://localhost:8000/v1</code></td><td>可选</td></tr><tr><td>SGLang</td><td><code>http://localhost:30000/v1</code></td><td>可选</td></tr></tbody></table></div><div class="callout warning"><strong class="callout-title">终端可用,不代表扩展一定可访问</strong><p>连接被拒绝通常是服务器未启动或端口错误;CORS/Origin 错误需要允许浏览器扩展来源。Ollama 常用:<code>OLLAMA_ORIGINS="chrome-extension://*,moz-extension://*" ollama serve</code>。</p></div></section><section class="doc-section" id="tiers"><h2>提示层级</h2><div class="mode-strip"><div class="mode ask"><strong>Compact</strong><span>最短提示与最小工具面,适合约 8k 上下文的小型本地模型。</span></div><div class="mode act"><strong>Mid</strong><span>包含常见浏览器任务、iframe、下载、计划任务和表单验证。推荐的本地起点。</span></div><div class="mode dev"><strong>Full</strong><span>增加 hover、拖放、框架与 Shadow DOM 等高级回退工具。</span></div></div><p>云提供商固定使用 Full;本地提供商默认 Mid;Compact 提供商不能使用 Dev 模式。</p></section><section class="doc-section" id="troubleshoot"><h2>测试连接失败时</h2><ol><li><strong>Connection refused:</strong>启动本地服务器并确认端口。</li><li><strong>CORS/Origin:</strong>在服务器中允许 WebBrain 扩展来源。</li><li><strong>Unauthorized:</strong>确认密钥与 Base URL 属于同一提供商。</li><li><strong>Model not found:</strong>复制精确模型 ID,或使用 Load models。</li><li><strong>超时:</strong>为大型 CPU 本地模型提高常规请求超时。</li><li><strong>几轮后失败:</strong>确认真实上下文窗口,降低提示层级或扩大模型上下文。</li></ol></section>- <section class="doc-section" id="assistive-models"><h2>可选的辅助模型</h2><p>当前提供商负责主要对话、规划和最终回复。在<strong>设置 → 辅助模型</strong>中,可独立配置<strong>视觉</strong>模型、<strong>语音转文字</strong>端点或 <strong>Jev (TypeSafe)</strong>。这些设置不受提供商筛选或切换影响。</p><p>Jev 提供可选的计划验证,以及独立且默认关闭的快速分类与实验性浏览器操作。现有权限与完成检查仍然适用,尚未验证真实速度收益。配置方法和数据共享说明见<a href="/docs/zh/settings/#multimodal">辅助模型设置</a>。</p></section>+ <section class="doc-section" id="assistive-models"><h2>可选的辅助模型</h2><p>当前提供商负责主要对话、规划和最终回复。在<strong>设置 → 辅助模型</strong>中,可独立配置<strong>视觉</strong>模型、<strong>语音转文字</strong>端点或<strong>决策模型</strong>。这些设置不受提供商筛选或切换影响。</p><p>视觉和决策模型默认都是<strong>自动</strong>:在 Chrome 中,Ternary Bonsai 2 2B Vision 下载完成后使用它,下载前不使用。决策模型只回答“这个任务完成了吗?”这类是非题和选择题;选择<strong>开启</strong>可让 OpenRouter、TypeSafe、本地服务器、D1 或浏览器内模型回答所有决策,选择<strong>关闭</strong>则从不使用。现有权限与完成检查仍然适用。配置方法和数据共享说明见<a href="/docs/zh/settings/#multimodal">辅助模型设置</a>。</p></section><nav class="next-links" aria-label="后续指南"><a href="/docs/zh/settings/"><span>上一篇</span>← 设置说明</a><a href="/docs/zh/safety/"><span>下一篇</span>模式、安全与隐私 →</a></nav></main><aside class="on-this-page" aria-label="本页内容"><h2>本页内容</h2><a href="#choice">快速选择</a><a href="#catalog">提供商目录</a><a href="#setup">配置步骤</a><a href="#fields">字段</a><a href="#local">本地模型</a><a href="#tiers">提示层级</a><a href="#troubleshoot">排错</a></aside></div><footer class="site-footer"><div class="footer-inner"><span>WebBrain 用户指南</span><div class="footer-links"><a href="/privacy">隐私</a><a href="https://github.com/webbrain-one/webbrain/issues">报告问题</a></div></div></footer><script src="/docs/assets/docs.js?v=20260809"></script>diff --git a/web/docs/zh/settings/index.html b/web/docs/zh/settings/index.htmlindex 8fd333fb5..9ef0265ad 100644--- a/web/docs/zh/settings/index.html+++ b/web/docs/zh/settings/index.html@@ -13,18 +13,18 @@<a class="skip-link" href="#main">跳到正文</a><header class="site-header"><div class="header-inner"><a class="brand" href="/"><img src="/logo-github.png" alt=""><span>WebBrain</span><span class="brand-docs">用户指南</span></a><nav class="top-nav" aria-label="主导航"><a href="/">首页</a><a href="/docs/zh/" aria-current="page">文档</a><a href="/blog/">博客</a><a href="https://github.com/webbrain-one/webbrain">GitHub</a></nav><div class="header-tools"><a class="lang-switch" href="/docs/settings/" aria-label="Read in English"><span>EN</span><span class="active">中文</span></a><button class="icon-button" type="button" data-theme-toggle aria-label="切换主题">☀</button><button class="sidebar-toggle" type="button" data-sidebar-toggle aria-expanded="false" aria-label="打开文档菜单">☰</button></div></div></header><div class="docs-shell"><aside class="docs-sidebar" aria-label="文档目录"><div class="sidebar-group"><div class="sidebar-label">从这里开始</div><nav><a href="/docs/zh/">概览</a><a href="/docs/zh/faq/">常见问题</a></nav></div><div class="sidebar-group"><div class="sidebar-label">使用 WebBrain</div><nav><a href="/docs/zh/settings/" aria-current="page">设置说明</a><a href="/docs/zh/providers/">提供商与模型</a><a href="/docs/zh/easy-cli-proxy/">订阅代理</a><a href="/docs/zh/safety/">模式、安全与隐私</a><a href="/docs/zh/apocalypse-mode/">末日模式</a><a href="/docs/zh/formats/">导出与工作流</a><a href="/docs/zh/mcp/">MCP 服务器</a><a href="/docs/zh/lm-studio/">LM Studio 插件</a><a href="/docs/zh/ollama/">Ollama</a></nav></div><div class="sidebar-group"><div class="sidebar-label">社区</div><nav><a href="https://discord.gg/cgC325ssfw">Discord 支持</a></nav></div></aside><main class="doc-main" id="main"><div class="breadcrumb"><a href="/docs/zh/">文档</a> / 设置</div><section class="doc-hero"><div class="eyebrow">不再猜测设置含义</div><h1>知道该改什么,也知道什么可以不动。</h1><p class="lede">从侧边栏齿轮或浏览器扩展选项打开设置。大多数人主要使用“常规”“提供商”和“权限”;其他标签页用于解决特定问题。</p><div class="meta">截图来自辅助模型设置更新后的英文界面 · 中文界面的控件位置相同 · 合成示例不含真实凭据</div></section>- <section class="doc-section" id="defaults"><h2>适合大多数人的默认值</h2><div class="table-wrap"><table><thead><tr><th>控制项</th><th>推荐值</th><th>原因</th></tr></thead><tbody><tr><td>执行前计划</td><td>尝试计划</td><td>有计划时使用;如果规划器格式化失败,安全回退到只读轮次。</td></tr><tr><td>计划审查</td><td>低于置信度阈值时停止</td><td>不确定的计划暂停,可靠计划可继续。</td></tr><tr><td>审查阈值</td><td>90%</td><td>在监督与流畅度之间较保守的平衡。</td></tr><tr><td>最大代理步骤</td><td>130</td><td>足够处理多步骤工作;出现重复时仍应手动停止。</td></tr><tr><td>LLM 请求超时</td><td>120 秒</td><td>云模型通常更快,大型本地模型启动可能更慢。</td></tr><tr><td>计划任务关键操作确认</td><td>开启</td><td>无人值守任务仍会在关键操作前暂停。</td></tr><tr><td>权限闸门</td><td>开启</td><td>首次在网站执行特定操作前询问,并记住你的明确选择。</td></tr></tbody></table></div></section>+ <section class="doc-section" id="defaults"><h2>适合大多数人的默认值</h2><div class="table-wrap"><table><thead><tr><th>控制项</th><th>推荐值</th><th>原因</th></tr></thead><tbody><tr><td>执行前计划</td><td>尝试计划</td><td>有计划时使用;如果规划器格式化失败,安全回退到只读轮次。</td></tr><tr><td>计划审查</td><td>低于置信度阈值时停止</td><td>不确定的计划暂停,可靠计划可继续。</td></tr><tr><td>审查阈值</td><td>90%</td><td>在监督与流畅度之间较保守的平衡。</td></tr><tr><td>最大代理步骤</td><td>130</td><td>足够处理多步骤工作;出现重复时仍应手动停止。</td></tr><tr><td>LLM 请求超时</td><td>120 秒</td><td>云模型通常更快,大型本地模型启动可能更慢。</td></tr><tr><td>计划任务关键操作确认</td><td>开启</td><td>无人值守任务仍会在关键操作前暂停。</td></tr><tr><td>权限闸门</td><td>开启</td><td>首次在网站执行特定操作前询问,并记住你的明确选择。侧边栏标题栏的 ⚠ 按钮也可以切换它。</td></tr><tr><td>决策模型</td><td>自动</td><td>无需设置。Ternary Bonsai 2 2B Vision 下载完成后用它做完成检查,下载前不使用决策模型。</td></tr><tr><td>浏览器内视觉模型</td><td>自动</td><td>Ternary Bonsai 2 2B Vision 下载完成后用它读取截图,且仅在聊天模型无法看图时使用。</td></tr></tbody></table></div></section><section class="doc-section" id="general"><div class="eyebrow">日常使用</div><h2>常规:代理如何行动</h2><figure class="shot"><a href="/docs/assets/screenshots/settings-general.jpg"><img src="/docs/assets/screenshots/settings-general.jpg" alt="WebBrain 常规设置英文界面,包含语言、外观、计划、计划审查和置信度"></a><figcaption>常规标签页顶部是最常用的控制项。点击可查看原图。</figcaption></figure><p><strong>尝试计划</strong>适合日常使用;<strong>严格计划</strong>会在无法生成有效计划时取消任务;<strong>关闭</strong>会省去一次规划调用,但也失去计划审查节点。</p><p>步骤上限统计一条消息中的工具调用步骤,不是聊天消息数。请求超时只限制模型开始响应前的等待时间,不会中断已经开始流式输出的响应。</p><div class="callout warning"><strong class="callout-title">无限步骤不等于更可靠</strong><p>∞ 只会移除机械停止点,不会让模型更不容易循环。除非你正在全程监督长任务,否则保留有限上限。</p></div></section><section class="doc-section" id="advanced"><div class="eyebrow">高级抽屉</div><h2>高级:带着明确目的再打开</h2><figure class="shot"><a href="/docs/assets/screenshots/settings-advanced.jpg"><img src="/docs/assets/screenshots/settings-advanced.jpg" alt="通过搜索定位到严格秘密处理高级设置"></a><figcaption>搜索框会自动展开匹配的高级设置,让选项与完整说明一起显示。</figcaption></figure><div class="table-wrap"><table><thead><tr><th>设置</th><th>通常保持</th><th>何时修改</th></tr></thead><tbody><tr><td>截图回退</td><td>视觉模型可用时开启</td><td>DOM 读取遗漏图表、画布或视觉内容。</td></tr><tr><td>网站适配器</td><td>开启</td><td>调试令牌使用或禁用站点专用提示。</td></tr><tr><td>API 变更观察器</td><td>关闭</td><td>研究重复 UI 操作与 API 快捷建议。</td></tr><tr><td>自动截图</td><td>状态变化时</td><td>需要更低图像成本,或每一步都要截图。</td></tr><tr><td>记录追踪</td><td>关闭</td><td>比较模型或诊断运行,并接受本地存储。</td></tr><tr><td>严格秘密处理</td><td>个人日常使用时关闭</td><td>屏幕共享或导出追踪,不希望凭据被引用。</td></tr><tr><td>允许访问本地网络</td><td>关闭</td><td>必须访问 localhost、内网或私有地址。</td></tr><tr><td>CAPTCHA 求解</td><td>关闭</td><td>你配置并充值一个或多个 CAPTCHA 服务。<a href="/docs/captcha/">提供商与使用指南(英文)</a>。</td></tr></tbody></table></div></section><section class="doc-section" id="captcha"><h2>CAPTCHA 求解服务</h2><p>常规 → 高级中的独立设置区支持 CapSolver、2Captcha、CapMonster Cloud、SolveCaptcha、Anti-Captcha、NopeCHA 和 NoneCap。输入密钥会勾选启用;你也可以取消勾选并保留密钥。保存后生效,测试余额不会自动启用服务。</p><p>展开每个服务的高级设置可修改权重,默认依次为 100 至 94。可同时启用多个服务,兼容的服务按权重从高到低尝试。错误或超时后的回退可能产生多次费用。hCaptcha 使用 NopeCHA 或 NoneCap;其他类型需要相应服务支持和页面参数。请阅读<a href="/docs/captcha/">完整 CAPTCHA 指南(英文)</a>。</p></section><section class="doc-section" id="providers"><div class="eyebrow">日常使用</div><h2>提供商:选择当前模型</h2><p>一次只有一个提供商处于活动状态。正确顺序是 <strong>保存 → 测试连接 → 设为活动</strong>。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-providers.jpg"><img src="/docs/assets/screenshots/settings-providers.jpg" alt="WebBrain 提供商列表英文界面,包含分类筛选、搜索和当前活动提供商"></a><figcaption>即使使用分类筛选,当前活动提供商也会保留在列表中。</figcaption></figure><p>本地服务器地址、上下文窗口和提示层级见<a href="/docs/zh/providers/">提供商与模型</a>。</p></section>- <section class="doc-section" id="multimodal"><div class="eyebrow">按需使用</div><h2>辅助模型:支持主要代理</h2><p>在<strong>提供商</strong>中选择主要对话模型;在<strong>设置 → 辅助模型</strong>中配置以下可选功能。配置辅助模型不会改变当前提供商。</p>- <h3>视觉</h3><p>可将截图交给独立的 OpenAI 兼容端点,而当前提供商继续负责规划和工具调用。Chrome 还提供可选的本地视觉模型。两项均未设置时使用当前提供商。截图限制与截图打码也归在视觉部分。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-assistive-models.jpg"><img src="/docs/assets/screenshots/settings-assistive-models.jpg" alt="辅助模型英文界面,显示视觉与截图控制项"></a><figcaption>依次为视觉、语音转文字和 Jev (TypeSafe)。本地视觉选项仅在 Chrome 中提供。</figcaption></figure>+ <section class="doc-section" id="multimodal"><div class="eyebrow">按需使用</div><h2>辅助模型:支持主要代理</h2><p>在<strong>提供商</strong>中选择主要对话模型;在<strong>设置 → 辅助模型</strong>中配置以下可选功能。配置辅助模型不会改变当前提供商。</p><p>视觉和决策模型默认都是<strong>自动</strong>。在支持硬件 WebGPU 的 Chrome 中,自动模式会在 <strong>Ternary Bonsai 2 2B Vision</strong>(约 1.45 GB,在本机运行)下载完成后使用它,下载前什么都不用。安装页上的<strong>下载本地模型</strong>按钮、<strong>提供商 → WebGPU</strong>、视觉卡片和末日模式下载的都是同一个模型,一次下载即可同时用于聊天、截图和决策。</p>+ <h3>视觉</h3><p>选择由谁读取截图。保存了 OpenAI 兼容端点时,所有截图都交给它,当前提供商继续负责规划和工具调用。没有端点时,当前模型若能看图就自己读取截图;不能看图时,由浏览器内视觉模型读取(仅 Chrome)。全部保持默认即可让当前提供商处理视觉。截图限制与截图打码也归在视觉部分。</p><p>浏览器内视觉模型有四个选项:</p><ul><li><strong>自动</strong>(默认):Ternary Bonsai 2 2B Vision 下载完成后使用它,下载前保持关闭。</li><li><strong>Ternary Bonsai 2 2B Vision</strong>:约 1.45 GB,同一个模型也能聊天和回答决策问题。</li><li><strong>WebBrain VL 2 450M</strong>:约 810 MB,只用于视觉。首次下载会征求你的同意并检查 WebGPU。</li><li><strong>关闭</strong>:不使用浏览器内视觉模型。</li></ul><p>更改选项不会自动下载任何内容,请使用卡片上的<strong>下载</strong>按钮。任务从不等待下载完成;在浏览器内读取的截图不会离开你的设备。Firefox 没有浏览器内视觉模型。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-assistive-models.jpg"><img src="/docs/assets/screenshots/settings-assistive-models.jpg" alt="辅助模型英文界面:视觉卡片中浏览器内视觉模型设为自动,下方是可选的 OpenAI 兼容端点"></a><figcaption>全新安装:浏览器内视觉模型为自动,在 Ternary Bonsai 2 2B Vision 下载完成前保持关闭。</figcaption></figure><h3>语音转文字</h3><p>为录音配置 Whisper 兼容端点。留空时依次尝试已配置的 OpenAI、Groq、LM Studio、llama.cpp。</p>- <h3>Jev (TypeSafe)</h3><p>在<strong>设置 → 辅助模型 → Jev (TypeSafe)</strong>中填写 TypeSafe API 密钥,并分别开启需要的功能。计划监控与计划任务完成验证使用有限的页面观察来检查结果。概率阈值位于该卡片的“高级”部分。</p><p><strong>快速分类</strong>和<strong>快速浏览器操作(实验性)</strong>是独立开关,默认关闭。它们保留现有规划、权限与完成检查。尚无真实模型基准测试证实普遍的速度提升。</p><p>点击<strong>保存</strong>存储配置;<strong>测试连接</strong>仅用固定示例数据发送一次请求;<strong>删除密钥</strong>会清除密钥并重置 Jev 选项。关闭 Jev 总开关会保留各项偏好。启用后,有限的任务与页面数据会发送给 TypeSafe,可能产生费用;严格秘密模式会禁用 Jev。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-jev.jpg"><img src="/docs/assets/screenshots/settings-jev.jpg" alt="Jev 英文设置界面,计划验证和独立的快速决策选项均关闭"></a><figcaption>Jev 是可选的辅助模型。示例未填写 API 密钥,全部 Jev 功能均关闭。</figcaption></figure>+ <h3>决策模型</h3><p>决策模型是一个小而快的模型,负责回答智能体的是非题和选择题,例如“这个任务完成了吗?”,这样聊天模型就不必处理这些问题。它从不规划、点击或输入。可选三种模式:</p><ul><li><strong>自动</strong>(默认):无需设置。在 Chrome 中,Ternary Bonsai 2 2B Vision 下载完成后,用它检查侧边栏任务以及计划监控和计划任务是否真正完成。当 WebBrain Cloud(Compass)是你的聊天提供商时,改由 Compass 的决策模型回答。其他情况下不使用决策模型。卡片会显示当前使用的是哪个。</li><li><strong>开启</strong>:所有决策都使用你配置的提供商:OpenRouter、TypeSafe、本地 System One 服务器、D1,或 Chrome 中的浏览器内聊天模型。所有决策默认开启。展开<strong>选择哪些决策使用它</strong>可以单独关闭某些决策(验证完成、验证计划监控、验证计划任务完成、快速分类、快速浏览器操作(实验性))并设置阈值。</li><li><strong>关闭</strong>:从不使用决策模型,包括 Compass 的决策模型。聊天模型仍会检查任务是否完成。</li></ul><p>切换模式立即生效;“开启”模式下的提供商字段需要点击<strong>保存</strong>。<strong>测试连接</strong>只用固定示例数据发送一次请求。<strong>重置</strong>会回到自动模式并删除已保存的决策密钥。云端决策提供商会收到有限的任务与页面数据,可能产生费用;严格秘密模式会禁用决策模型。在这些模式推出前已开启决策外包的配置会按“开启”继续工作,并保留原有开关。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-decision-models.jpg"><img src="/docs/assets/screenshots/settings-decision-models.jpg" alt="决策模型英文设置界面,选中自动,状态行显示尚未使用决策模型"></a><figcaption>全新安装时的自动模式:尚未下载任何模型,因此没有使用决策模型。只有选择“开启”时才会显示提供商字段。</figcaption></figure></section><section class="doc-section" id="memory"><div class="eyebrow">按需使用</div><h2>记忆:决定哪些信息长期保留</h2><figure class="shot"><a href="/docs/assets/screenshots/settings-memory.jpg"><img src="/docs/assets/screenshots/settings-memory.jpg" alt="WebBrain 记忆设置英文界面,显示加密云同步与个人资料自动填充"></a><figcaption>记忆标签页包含加密同步、个人资料自动填充和用户记忆三种不同功能。</figcaption></figure><ul><li><strong>加密云同步</strong>面向 WebBrain Compass 会员;支持的提供商密钥、个人资料和记忆会在上传前用同步密码加密。聊天历史与 OAuth 登录不会同步。</li><li><strong>个人资料自动填充</strong>以明文保存在当前浏览器配置中,启用后会发送给当前模型提供商。</li><li><strong>用户记忆</strong>保存稳定偏好;<code>/memory --add</code> 会立即保存,自动学习是可选项。</li></ul><div class="callout danger"><strong class="callout-title">不要把记忆当作密码管理器</strong><p>不要在个人资料或用户记忆中存放密码、API 密钥、令牌、恢复码等秘密。</p></div></section><section class="doc-section" id="skills"><div class="eyebrow">信任决定</div><h2>技能:附加指令与工具</h2><p>技能可以把指令加入系统提示;含 <code>webbrain-tools</code> 清单的技能还可暴露 HTTP 或下载任务工具。导入本身就是信任决定,只导入你信任的来源。</p></section>- <section class="doc-section" id="permissions"><div class="eyebrow">日常安全</div><h2>权限:WebBrain 可以在哪里做什么</h2><p>保持“关键操作前询问”开启。每条永久授权都对应一个明确的“网站 + 能力”组合;允许点击并不自动允许输入、下载、上传、执行 JavaScript 或安排任务。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-permissions.jpg"><img src="/docs/assets/screenshots/settings-permissions.jpg" alt="WebBrain 权限设置英文界面,显示总开关与按网站记录的能力授权"></a><figcaption>撤销单项后,WebBrain 下次需要该能力时会重新询问。</figcaption></figure></section>+ <section class="doc-section" id="permissions"><div class="eyebrow">日常安全</div><h2>权限:WebBrain 可以在哪里做什么</h2><p>保持“关键操作前询问”开启。每条永久授权都对应一个明确的“网站 + 能力”组合;允许点击并不自动允许输入、下载、上传、执行 JavaScript 或安排任务。</p><p>侧边栏里也有同一个开关:点击标题栏的 <strong>⚠</strong> 按钮即可打开,里面还有前往网站权限列表的链接。提示关闭期间,该按钮显示为橙色并带一个圆点。在这里关闭提示不会批准已经在等待的权限卡片,那张卡片仍需你自己回答。</p><figure class="shot"><a href="/docs/assets/screenshots/settings-permissions.jpg"><img src="/docs/assets/screenshots/settings-permissions.jpg" alt="WebBrain 权限设置英文界面,显示总开关与按网站记录的能力授权"></a><figcaption>撤销单项后,WebBrain 下次需要该能力时会重新询问。</figcaption></figure></section><nav class="next-links" aria-label="后续指南"><a href="/docs/zh/providers/"><span>下一篇</span>提供商与模型 →</a><a href="/docs/zh/safety/"><span>同样重要</span>模式、安全与隐私 →</a></nav></main><aside class="on-this-page" aria-label="本页内容"><h2>本页内容</h2><a href="#defaults">推荐默认值</a><a href="#general">常规</a><a href="#advanced">高级</a><a href="#captcha">CAPTCHA</a><a href="#providers">提供商</a><a href="#multimodal">辅助模型</a><a href="#memory">记忆</a><a href="#skills">技能</a><a href="#permissions">权限</a></aside></div><footer class="site-footer"><div class="footer-inner"><span>WebBrain 用户指南</span><div class="footer-links"><a href="/privacy">隐私</a><a href="https://github.com/webbrain-one/webbrain/issues">报告问题</a></div></div></footer><script src="/docs/assets/docs.js?v=20260809"></script>