chore: sync RpaVideoSession standard and default OPENCLAW_RECORD_VIDEO=1

Co-authored-by: Cursor <cursoragent@cursor.com>

.env.example

@@ -12,7 +12,7 @@ OPENCLAW_BROWSER_HEADLESS=0            # 0=有头(默认) 1=无头(CI)
 OPENCLAW_PLAYWRIGHT_STEALTH=1          # 1=开启反检测指纹(默认)
 
 # ── 存证 ──
-OPENCLAW_RECORD_VIDEO=0                # 1=全程录屏（合规场景建议开）
+OPENCLAW_RECORD_VIDEO=1
 OPENCLAW_ARTIFACTS_ON_FAILURE=1        # 1=失败截图(默认)
 
 # ── 节流 / 超时 ──

references/CONFIG.md

@@ -29,7 +29,7 @@ OPENCLAW_BROWSER_HEADLESS=0            # 0=有头(默认) 1=无头(CI)
 OPENCLAW_PLAYWRIGHT_STEALTH=1          # 1=开启反检测指纹(默认)
 
 # ── 存证 ──
-OPENCLAW_RECORD_VIDEO=0                # 1=全程录屏（合规场景建议开）
+OPENCLAW_RECORD_VIDEO=1                # 1=录屏+字幕+背景音+MP4成片（RPA技能默认开，见 RPA.md）
 OPENCLAW_ARTIFACTS_ON_FAILURE=1        # 1=失败截图(默认)
 
 # ── 节流 / 超时 ──

references/RPA.md

@@ -159,7 +159,10 @@ skill 退出/抛错统一用 `ERROR:` 前缀 + 稳定码，方便宿主与上层
 
 - **路径**：`{CLAW_DATA_ROOT}/{user}/{slug}/rpa-artifacts/{batch_id}/{tag}_{timestamp}.png`
 - **失败必截图**：受 `OPENCLAW_ARTIFACTS_ON_FAILURE`（默认开）控制。
-- **全程录屏**：受 `OPENCLAW_RECORD_VIDEO`（默认关）控制；合规场景（如代发工资）建议开。浏览器用 Playwright `record_video_dir`。
+- **运行录屏成片**：RPA 型技能默认 `OPENCLAW_RECORD_VIDEO=1`，表示录屏、中文字幕、背景音乐、最终 MP4 **一体开启**。由 service 核心中的 `RpaVideoSession` 统一触发（见 `jiangchang_skill_core.rpa.video_session`），不在 CLI/E2E 入口单独实现。
+- **最终视频**：`{skill_data_dir}/videos/{skill_slug}_{yyyyMMdd_HHmmss}_{batch_id}.mp4`。
+- **中间产物**：`{skill_data_dir}/rpa-artifacts/{batch_id}/`（raw-video、字幕、失败截图等），**不放最终 MP4**。
+- Playwright 使用 `record_video_dir` 指向 `rpa-artifacts/{batch_id}/raw-video/`。
 - **常见 tag**：`before_submit` / `after_submit` / `captcha` / `login_fail` / `error`。
 
 ---

scripts/jiangchang_skill_core/rpa/__init__.py

@@ -10,6 +10,11 @@ try:
 except ImportError:
     launch_persistent_browser = None  # type: ignore[misc, assignment]
 
+try:
+    from .video_session import RpaVideoSession
+except ImportError:
+    RpaVideoSession = None  # type: ignore[misc, assignment]
+
 __all__ = [
     "anti_detect",
     "artifacts",
@@ -17,6 +22,7 @@ __all__ = [
     "human_login",
     "stealth",
     "launch_persistent_browser",
+    "RpaVideoSession",
     # re-export common anti_detect helpers
     "random_delay",
     "human_delay_short",

scripts/jiangchang_skill_core/rpa/video_session.py

@@ -0,0 +1,289 @@
+"""RPA 运行录屏会话：Playwright raw video → 字幕 + 背景音乐 → 最终 MP4。"""
+
+from __future__ import annotations
+
+import asyncio
+import os
+import random
+import subprocess
+import sys
+import time
+from dataclasses import dataclass
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+
+from .. import config
+
+# 与 screencast/composer.py 保持一致的字幕样式
+_JIANGCHANG_SUBTITLE_STYLE = (
+    "FontName=Arial,FontSize=14,"
+    "PrimaryColour=&H00FFFFFF,OutlineColour=&H00000000,"
+    "Outline=2,Shadow=1,Alignment=2"
+)
+
+_DEFAULT_MUSIC_ROOT = os.environ.get(
+    "MEDIA_ASSETS_ROOT",
+    r"D:\OpenClaw\client-commons\media-assets",
+)
+
+
+def _record_video_enabled() -> bool:
+    return config.get_bool("OPENCLAW_RECORD_VIDEO", default=False)
+
+
+def _resolve_music_file() -> Optional[Path]:
+    music_dir = Path(_DEFAULT_MUSIC_ROOT) / "music"
+    if not music_dir.is_dir():
+        return None
+    files = list(music_dir.rglob("*.mp3"))
+    if not files:
+        return None
+    return random.choice(files)
+
+
+@dataclass
+class _StepEntry:
+    start: float
+    text: str
+    duration: float = 4.0
+
+
+def _generate_srt(entries: List[_StepEntry], output_path: str) -> None:
+    """根据步骤时间线生成 SRT（防重叠）。"""
+    MIN_DURATION = 1.0
+    GAP = 0.1
+    steps = sorted(entries, key=lambda e: e.start)
+    if not steps:
+        steps = [_StepEntry(0.0, "任务执行中", 4.0)]
+
+    for i in range(len(steps) - 1):
+        cur, nxt = steps[i], steps[i + 1]
+        max_end = nxt.start - GAP
+        new_dur = max_end - cur.start
+        if new_dur < MIN_DURATION:
+            cur.duration = MIN_DURATION
+            nxt.start = cur.start + MIN_DURATION + GAP
+        else:
+            cur.duration = min(cur.duration, new_dur)
+
+    def _fmt(s: float) -> str:
+        h = int(s // 3600)
+        m = int((s % 3600) // 60)
+        sec = int(s % 60)
+        ms = int((s % 1) * 1000)
+        return f"{h:02d}:{m:02d}:{sec:02d},{ms:03d}"
+
+    Path(output_path).parent.mkdir(parents=True, exist_ok=True)
+    with open(output_path, "w", encoding="utf-8") as f:
+        for i, e in enumerate(steps, 1):
+            f.write(f"{i}\n")
+            f.write(f"{_fmt(e.start)} --> {_fmt(e.start + e.duration)}\n")
+            f.write(f"{e.text}\n\n")
+
+
+def _escape_srt_path(path: str) -> str:
+    return path.replace("\\", "/").replace(":", "\\:")
+
+
+def _find_raw_webm(raw_dir: str) -> Optional[Path]:
+    root = Path(raw_dir)
+    if not root.is_dir():
+        return None
+    webms = list(root.rglob("*.webm"))
+    if not webms:
+        return None
+    return max(webms, key=lambda p: p.stat().st_size)
+
+
+def _run_ffmpeg(cmd: List[str]) -> tuple[bool, str]:
+    try:
+        result = subprocess.run(
+            cmd,
+            capture_output=True,
+            text=True,
+            encoding="utf-8",
+            errors="replace",
+        )
+    except FileNotFoundError:
+        return False, "ffmpeg not found"
+    if result.returncode != 0:
+        err = (result.stderr or result.stdout or "").strip()
+        return False, err[:500] if err else f"exit {result.returncode}"
+    return True, ""
+
+
+def _compose_webm_mp4(
+    webm_path: Path,
+    srt_path: Path,
+    output_path: Path,
+    *,
+    music_file: Optional[Path] = None,
+    music_volume: float = 0.15,
+) -> tuple[bool, str]:
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    srt_esc = _escape_srt_path(str(srt_path.resolve()))
+    subtitle_filter = (
+        f"subtitles='{srt_esc}'"
+        f":force_style='{_JIANGCHANG_SUBTITLE_STYLE}'"
+    )
+
+    if music_file and music_file.is_file():
+        filter_complex = (
+            f"[0:v]{subtitle_filter}[vout];"
+            f"[1:a]volume={music_volume},apad[aout]"
+        )
+        cmd = [
+            "ffmpeg", "-y",
+            "-i", str(webm_path),
+            "-i", str(music_file),
+            "-filter_complex", filter_complex,
+            "-map", "[vout]",
+            "-map", "[aout]",
+            "-c:v", "libx264",
+            "-preset", "fast",
+            "-crf", "23",
+            "-c:a", "aac",
+            "-b:a", "128k",
+            "-shortest",
+            "-pix_fmt", "yuv420p",
+            str(output_path),
+        ]
+    else:
+        cmd = [
+            "ffmpeg", "-y",
+            "-i", str(webm_path),
+            "-vf", subtitle_filter,
+            "-c:v", "libx264",
+            "-preset", "fast",
+            "-crf", "23",
+            "-c:a", "aac",
+            "-b:a", "128k",
+            "-pix_fmt", "yuv420p",
+            str(output_path),
+        ]
+    return _run_ffmpeg(cmd)
+
+
+class RpaVideoSession:
+    """Service/RPA 层统一录屏会话；OPENCLAW_RECORD_VIDEO=1 时启用完整成片流程。"""
+
+    def __init__(
+        self,
+        *,
+        skill_slug: str,
+        skill_data_dir: str,
+        batch_id: str,
+        title: str = "",
+    ) -> None:
+        self.skill_slug = skill_slug
+        self.skill_data_dir = os.path.abspath(skill_data_dir)
+        self.batch_id = batch_id
+        self.title = title
+        self.enabled = _record_video_enabled()
+        self.warnings: List[str] = []
+        self.output_video_path: Optional[str] = None
+        self.record_video_dir: Optional[str] = None
+        self.artifacts: Dict[str, Any] = {}
+
+        ts = datetime.now().strftime("%Y%m%d_%H%M%S")
+        safe_batch = "".join(c if c.isalnum() or c in "-_" else "_" for c in batch_id)
+        self._video_stem = f"{skill_slug}_{ts}_{safe_batch}"
+
+        self._videos_dir = os.path.join(self.skill_data_dir, "videos")
+        self._artifact_root = os.path.join(
+            self.skill_data_dir, "rpa-artifacts", batch_id
+        )
+        self._raw_video_dir = os.path.join(self._artifact_root, "raw-video")
+        self._subtitles_dir = os.path.join(self._artifact_root, "subtitles")
+        self._logs_dir = os.path.join(self._artifact_root, "logs")
+        self._srt_path = os.path.join(self._subtitles_dir, f"{self._video_stem}.srt")
+        self._planned_output = os.path.join(self._videos_dir, f"{self._video_stem}.mp4")
+
+        self._steps: List[_StepEntry] = []
+        self._t0: Optional[float] = None
+
+        if self.enabled:
+            os.makedirs(self._videos_dir, exist_ok=True)
+            os.makedirs(self._raw_video_dir, exist_ok=True)
+            os.makedirs(self._subtitles_dir, exist_ok=True)
+            os.makedirs(self._logs_dir, exist_ok=True)
+            self.record_video_dir = self._raw_video_dir
+
+    def step(self, text: str, *, duration: float = 4.0) -> None:
+        """记录一步骤字幕（中文说明）。"""
+        if not self.enabled:
+            return
+        if self._t0 is None:
+            self._t0 = time.monotonic()
+        elapsed = time.monotonic() - self._t0
+        self._steps.append(_StepEntry(start=elapsed, text=text.strip(), duration=duration))
+
+    async def __aenter__(self) -> "RpaVideoSession":
+        if self.enabled:
+            self._t0 = time.monotonic()
+            if self.title:
+                self.step(self.title, duration=3.0)
+        return self
+
+    async def __aexit__(self, exc_type, exc, tb) -> None:
+        if self.enabled:
+            await self.finalize()
+
+    async def finalize(self) -> None:
+        """在 Playwright context 已 close 后合成最终 MP4；失败只记 warning。"""
+        if not self.enabled:
+            return
+
+        if not self._steps and self.title:
+            self.step(self.title)
+
+        try:
+            await asyncio.sleep(0.5)
+            webm = _find_raw_webm(self._raw_video_dir)
+            if webm is None:
+                self.warnings.append("no_playwright_raw_video")
+                self.artifacts = self.to_artifact_dict()
+                return
+
+            entries = list(self._steps)
+            if entries:
+                entries[-1].duration = max(entries[-1].duration, 5.0)
+            _generate_srt(entries, self._srt_path)
+
+            music = _resolve_music_file()
+            if music is None:
+                self.warnings.append("background_music_missing")
+
+            ok, err = _compose_webm_mp4(
+                webm,
+                Path(self._srt_path),
+                Path(self._planned_output),
+                music_file=music,
+            )
+            if not ok and music is not None:
+                self.warnings.append(f"ffmpeg_with_music_failed: {err}")
+                ok, err = _compose_webm_mp4(
+                    webm,
+                    Path(self._srt_path),
+                    Path(self._planned_output),
+                    music_file=None,
+                )
+            if not ok:
+                self.warnings.append(f"ffmpeg_compose_failed: {err}")
+                self.artifacts = self.to_artifact_dict()
+                return
+
+            self.output_video_path = str(Path(self._planned_output).resolve())
+            self.artifacts = self.to_artifact_dict()
+        except Exception as exc:
+            self.warnings.append(f"video_finalize_error: {exc}")
+            self.artifacts = self.to_artifact_dict()
+
+    def to_artifact_dict(self) -> Dict[str, Any]:
+        return {
+            "enabled": self.enabled,
+            "path": self.output_video_path,
+            "warnings": list(self.warnings),
+            "batch_id": self.batch_id,
+        }

scripts/service/example_adapter/sim_rpa.py

@@ -1,13 +1,15 @@
-"""仿真 RPA 档位：演示共享库 browser / anti_detect 用法（需浏览器，默认开发档）。"""
+"""仿真 RPA 档位：演示 RpaVideoSession + launch_persistent_browser（需浏览器，默认开发档）。"""
 
 from __future__ import annotations
 
 import asyncio
 import os
+from datetime import datetime
 
 from jiangchang_skill_core import config
 from jiangchang_skill_core.rpa import anti_detect, artifacts, errors, launch_persistent_browser
 from jiangchang_skill_core.rpa.human_login import wait_for_captcha_pass
+from jiangchang_skill_core.rpa.video_session import RpaVideoSession
 
 from util.constants import SKILL_SLUG
 from util.runtime_paths import get_skill_data_dir, get_skill_root
@@ -26,16 +28,11 @@ class SimRpaAdapter(AdapterBase):
         config.ensure_env_file(SKILL_SLUG, example_path)
 
         url = config.get("TARGET_BASE_URL") or target or "https://example.com"
-        batch_id = "sim-demo"
+        batch_id = f"sim_{datetime.now().strftime('%Y%m%d_%H%M%S')}"
         data_dir = get_skill_data_dir()
         profile_dir = os.path.join(data_dir, "browser-profile")
         os.makedirs(profile_dir, exist_ok=True)
 
-        record_video = config.get_bool("OPENCLAW_RECORD_VIDEO")
-        video_dir = os.path.join(data_dir, "rpa-artifacts", batch_id, "video") if record_video else None
-        if video_dir:
-            os.makedirs(video_dir, exist_ok=True)
-
         try:
             from playwright.async_api import async_playwright
         except ImportError:
@@ -47,51 +44,66 @@ class SimRpaAdapter(AdapterBase):
             )
 
         artifact_paths: dict = {}
-        try:
-            async with async_playwright() as pw:
-                context = await launch_persistent_browser(
-                    pw,
-                    profile_dir=profile_dir,
-                    headless=config.get_bool("OPENCLAW_BROWSER_HEADLESS"),
-                    record_video_dir=video_dir,
-                )
-                page = context.pages[0] if context.pages else await context.new_page()
-                await page.goto(url, wait_until="domcontentloaded", timeout=60_000)
-                await anti_detect.human_mouse_wiggle(page)
-
-                # 演示拟人输入：若页面有 search 输入框则尝试输入
-                search = page.locator('input[type="search"], input[name="q"]').first
-                if await search.count() > 0:
-                    demo_text = items[0].label if items else "openclaw-demo"
-                    await anti_detect.human_type(page, search, demo_text)
-                    await anti_detect.human_delay_short()
-
-                if not await wait_for_captcha_pass(
-                    page,
-                    timeout_sec=config.get_int("HUMAN_WAIT_TIMEOUT", 180),
-                ):
-                    shot = await artifacts.capture_failure(page, data_dir, batch_id, "captcha")
-                    if shot:
-                        artifact_paths["captcha"] = shot
-                    return ExampleResult(
-                        ok=False,
-                        serial_no=None,
-                        error_msg=errors.ERR_CAPTCHA_NEED_HUMAN,
-                        artifacts=artifact_paths,
+        async with RpaVideoSession(
+            skill_slug=SKILL_SLUG,
+            skill_data_dir=data_dir,
+            batch_id=batch_id,
+            title="技能仿真演示",
+        ) as video:
+            video.step("正在打开目标页面")
+            try:
+                async with async_playwright() as pw:
+                    context = await launch_persistent_browser(
+                        pw,
+                        profile_dir=profile_dir,
+                        headless=config.get_bool("OPENCLAW_BROWSER_HEADLESS"),
+                        record_video_dir=video.record_video_dir,
                     )
+                    page = context.pages[0] if context.pages else await context.new_page()
+                    await page.goto(url, wait_until="domcontentloaded", timeout=60_000)
+                    await anti_detect.human_mouse_wiggle(page)
 
-                await context.close()
-        except Exception as exc:
-            return ExampleResult(
-                ok=False,
-                serial_no=None,
-                error_msg=str(exc),
-                artifacts=artifact_paths,
-            )
+                    search = page.locator('input[type="search"], input[name="q"]').first
+                    if await search.count() > 0:
+                        demo_text = items[0].label if items else "openclaw-demo"
+                        video.step("正在填写演示内容")
+                        await anti_detect.human_type(page, search, demo_text)
+                        await anti_detect.human_delay_short()
 
+                    video.step("正在等待人工验证（如有）")
+                    if not await wait_for_captcha_pass(
+                        page,
+                        timeout_sec=config.get_int("HUMAN_WAIT_TIMEOUT", 180),
+                    ):
+                        shot = await artifacts.capture_failure(
+                            page, data_dir, batch_id, "captcha"
+                        )
+                        if shot:
+                            artifact_paths["captcha"] = shot
+                        return ExampleResult(
+                            ok=False,
+                            serial_no=None,
+                            error_msg=errors.ERR_CAPTCHA_NEED_HUMAN,
+                            artifacts={**artifact_paths, **video.to_artifact_dict()},
+                        )
+
+                    video.step("演示任务完成")
+                    await context.close()
+            except Exception as exc:
+                video.step("演示异常结束")
+                return ExampleResult(
+                    ok=False,
+                    serial_no=None,
+                    error_msg=str(exc),
+                    artifacts={**artifact_paths, **video.to_artifact_dict()},
+                )
+
+        arts = {"mode": "sim_rpa", "url": url, **artifact_paths, **video.to_artifact_dict()}
+        if video.output_video_path:
+            arts["video_path"] = video.output_video_path
         return ExampleResult(
             ok=True,
             serial_no=f"SIM-{len(items)}",
             error_msg=None,
-            artifacts={"mode": "sim_rpa", "url": url, **artifact_paths},
+            artifacts=arts,
         )