openclaw
diff --git a/‎docs/cli/browser.md‎
Lines changed: 14 additions & 2 deletions b/‎docs/cli/browser.md‎
Lines changed: 14 additions & 2 deletions
diff --git a/‎docs/docs_map.md‎
Lines changed: 1 addition & 1 deletion b/‎docs/docs_map.md‎
Lines changed: 1 addition & 1 deletion
diff --git a/‎docs/tools/browser-control.md‎
Lines changed: 11 additions & 3 deletions b/‎docs/tools/browser-control.md‎
Lines changed: 11 additions & 3 deletions
diff --git a/‎docs/tools/browser.md‎
Lines changed: 7 additions & 0 deletions b/‎docs/tools/browser.md‎
Lines changed: 7 additions & 0 deletions
diff --git a/‎extensions/browser/index.test.ts‎
Lines changed: 2 additions & 0 deletions b/‎extensions/browser/index.test.ts‎
Lines changed: 2 additions & 0 deletions
diff --git a/‎extensions/browser/plugin-registration.ts‎
Lines changed: 3 additions & 0 deletions b/‎extensions/browser/plugin-registration.ts‎
Lines changed: 3 additions & 0 deletions
diff --git a/‎extensions/browser/skills/browser-automation/SKILL.md‎
Lines changed: 3 additions & 0 deletions b/‎extensions/browser/skills/browser-automation/SKILL.md‎
Lines changed: 3 additions & 0 deletions
@@ -135,7 +135,19 @@ openclaw browser close t1
 
 Raw target ids are volatile diagnostic handles, not durable agent memory: when Chromium replaces the underlying raw target during a navigation or form submit, OpenClaw keeps the stable `tabId`/label attached to the replacement tab when it can prove the match. Prefer `suggestedTargetId`.
 
-## Snapshot / screenshot / actions
+## Extract / snapshot / screenshot / actions
+
+Answer a question from the current page without printing the page content:
+
+```bash
+openclaw browser extract "What is the main conclusion?"
+openclaw browser extract "Which deadline is listed?" --target-id docs --timeout-ms 90000
+```
+
+`extract` uses the selected agent model, returns only the wrapped answer, and
+reports `NOT_FOUND` when the answer is absent. Its overall timeout defaults to
+60 seconds and is clamped to 5–120 seconds. It requires a Playwright-backed
+profile; use `snapshot` when you need refs or when extraction is unavailable.
 
 Snapshot:
 
@@ -275,7 +287,7 @@ Current existing-session limits:
 - File uploads require `--ref` / `--input-ref`, do not support CSS `--element`, and support one file at a time.
 - Dialog hooks do not support `--timeout`.
 - Screenshots support page captures and `--ref`, but not CSS `--element`.
-- `responsebody`, download interception, PDF export, and batch actions still require a managed browser or raw CDP profile.
+- `extract`, `responsebody`, download interception, PDF export, and batch actions still require a managed browser or raw CDP profile.
 
 ## Remote browser control (node host proxy)
 
 
@@ -1340,7 +1340,7 @@ Do not edit it by hand; run `pnpm docs:map:gen`.
   - H2: If the command is missing
   - H2: Profiles
   - H2: Tabs
-  - H2: Snapshot / screenshot / actions
+  - H2: Extract / snapshot / screenshot / actions
   - H2: State and storage
   - H2: Debugging
   - H2: Existing Chrome via MCP
 
@@ -23,7 +23,7 @@ agent tools, but nothing listens on the loopback control port.
 - Status/start/stop: `GET /`, `GET /doctor`, `POST /start`, `POST /stop`, `POST /reset-profile`
 - Profiles: `GET /profiles`, `POST /profiles/create`, `DELETE /profiles/:name`
 - Tabs: `GET /tabs`, `POST /tabs/open`, `POST /tabs/focus`, `DELETE /tabs/:targetId`, `POST /tabs/action`
-- Snapshot/screenshot: `GET /snapshot`, `POST /screenshot`
+- Snapshot/screenshot/extract: `GET /snapshot`, `POST /screenshot`, `POST /extract`
 - Actions: `POST /navigate`, `POST /act`
 - Hooks: `POST /hooks/file-chooser`, `POST /hooks/dialog`
 - Downloads: `POST /download`, `POST /wait/download`
@@ -84,8 +84,8 @@ Other runtime failures may still return `{ "error": "<message>" }` without a
 
 ### Playwright requirement
 
-Some features (navigate/act/AI snapshot/role snapshot, element screenshots,
-PDF) require Playwright. If Playwright isn't installed, those endpoints return
+Some features (navigate/act/AI snapshot/role snapshot, extract, element
+screenshots, PDF) require Playwright. If Playwright isn't installed, those endpoints return
 a clear 501 error.
 
 What still works without Playwright:
@@ -107,6 +107,7 @@ What still needs Playwright:
 - AI snapshots that depend on Playwright's native AI snapshot format
 - CSS-selector element screenshots (`--element`)
 - full browser PDF export
+- page-question extraction
 
 Element screenshots also reject `--full-page`; the route returns `fullPage is
 not supported for element screenshots`.
@@ -198,6 +199,7 @@ openclaw browser snapshot --urls
 openclaw browser snapshot --selector "#main" --interactive
 openclaw browser snapshot --frame "iframe#main" --interactive
 openclaw browser snapshot --out snapshot.txt
+openclaw browser extract "What is the page's main conclusion?"
 openclaw browser console --level error
 openclaw browser errors --clear
 openclaw browser requests --filter api --clear
@@ -265,6 +267,12 @@ openclaw browser set device "iPhone 14"
 
 Notes:
 
+- Use `browser extract "<question>"` or agent-tool `action="extract"` when you
+  need an answer from the current page but do not need interaction refs. It
+  sanitizes readable page content, caps it at 80,000 characters, runs one
+  model call, and returns only the wrapped answer. The overall timeout defaults
+  to 60 seconds and is clamped to 5–120 seconds. If extraction fails, fall back
+  to `snapshot`; existing-session profiles do not support extraction.
 - The agent-facing `browser` tool exposes `action=download` (required `ref` and
   `path`) and `action=waitfordownload` (optional `path`). Both return the saved
   download URL, suggested filename, and guarded local path. Explicit download
 
@@ -18,6 +18,7 @@ OpenClaw can run a **dedicated Chrome/Brave/Edge/Chromium profile** that the age
 - A separate browser profile named **openclaw** (orange accent by default).
 - Deterministic tab control (list/open/focus/close).
 - Agent actions (click/type/drag/select), snapshots, screenshots, PDFs.
+- Question answering over readable page text without returning a full snapshot.
 - Playwright-backed profiles save direct attachment navigations under the managed downloads directory and return `{ url, suggestedFilename, path }` metadata after final-URL policy validation.
 - Playwright-backed agent actions return a `downloads` array with the same managed metadata when the action immediately starts one or more downloads.
 - A bundled `browser-automation` skill that teaches agents the snapshot,
@@ -101,6 +102,12 @@ Plugin-bundled skills are listed in the agent's available skills when the
 plugin is enabled. The full skill instructions load on demand, so routine
 turns do not pay the full token cost.
 
+For “read this page and answer X,” use browser `action="extract"` with a
+`query`. It sends sanitized, bounded readable text through one model call and
+returns only the answer; keep `snapshot` for choosing actions and obtaining
+refs. Extraction requires a Playwright-backed profile and falls back to a
+snapshot workflow when it cannot complete.
+
 ## Missing browser command or tool
 
 If `openclaw browser` is unknown after an upgrade, `browser.request` is missing, or the agent reports the browser tool as unavailable, the usual cause is a `plugins.allow` list that omits `browser` and no root `browser` config block exists. Add it:
 
@@ -239,6 +239,7 @@ describe("browser plugin", () => {
 
     const tool = factory({
       sessionKey: "agent:main:webchat:direct:123",
+      agentId: "main",
       agentDir: "/tmp/agent",
       workspaceDir: "/tmp/workspace",
       activeModel: { provider: "openai", modelId: "gpt-5.5" },
@@ -251,6 +252,7 @@ describe("browser plugin", () => {
     await tool.execute("call-1", { action: "status" });
     expect(runtimeApiMocks.createBrowserTool).toHaveBeenCalledWith({
       agentSessionKey: "agent:main:webchat:direct:123",
+      agentId: "main",
       agentDir: "/tmp/agent",
       workspaceDir: "/tmp/workspace",
       activeModel: { provider: "openai", model: "gpt-5.5" },
 
@@ -63,6 +63,7 @@ function createLazyBrowserTool(opts?: {
   sandboxBridgeUrl?: string;
   allowHostControl?: boolean;
   agentSessionKey?: string;
+  agentId?: string;
   agentDir?: string;
   workspaceDir?: string;
   activeModel?: {
@@ -105,6 +106,7 @@ function createBrowserToolOptions(ctx: OpenClawPluginToolContext): {
   sandboxBridgeUrl?: string;
   allowHostControl?: boolean;
   agentSessionKey?: string;
+  agentId?: string;
   agentDir?: string;
   workspaceDir?: string;
   activeModel?: {
@@ -126,6 +128,7 @@ function createBrowserToolOptions(ctx: OpenClawPluginToolContext): {
       ? { allowHostControl: ctx.browser.allowHostControl }
       : {}),
     ...(ctx.sessionKey ? { agentSessionKey: ctx.sessionKey } : {}),
+    ...(ctx.agentId ? { agentId: ctx.agentId } : {}),
     ...(ctx.agentDir ? { agentDir: ctx.agentDir } : {}),
     ...(ctx.workspaceDir ? { workspaceDir: ctx.workspaceDir } : {}),
     ...(ctx.activeModel?.provider || ctx.activeModel?.modelId
 
@@ -21,6 +21,9 @@ Use this skill when you need the `browser` tool for anything beyond a single pag
    - `suggestedTargetId` is the label when one exists, otherwise the stable `tabId` handle like `t1`.
    - Avoid relying on raw DevTools `targetId` except for immediate diagnostics; it can change under Chromium target replacement.
 3. Read before you click:
+   - For “read the page and answer X,” use `action="extract"` with `query` so only the answer returns.
+   - Use `action="snapshot"` instead when you need action refs or page structure.
+   - If extract returns `NOT_FOUND` or asks for snapshot fallback, inspect the page with a snapshot.
    - Use `action="snapshot"` on the intended `targetId`.
    - Use the same `targetId` for follow-up actions so refs stay on the same tab.
    - For durable Playwright refs, request `refs="aria"` when supported. If you receive `axN` refs from `snapshotFormat="aria"`, use them only after that same snapshot call; stale or unbound `axN` refs fail fast and need a fresh snapshot.