@marshulll/openclaw-wecom 0.1.22 → 0.1.24

package/README.en.md

@@ -93,10 +93,12 @@ Install guide: `docs/INSTALL.md`
 - `/sendfile`: send files from server (multiple absolute paths)
   - Directories are zipped automatically
   - Example: `/sendfile /tmp/openclaw-wecom /home/shu/Desktop/report.pdf`
+  - Natural language also works: "send me this file image-xxx.jpg" (default match in `media.tempDir`)
 
 ## Media auto recognition (optional)
 - **Voice send/receive does NOT require API**; only auto transcription needs an OpenAI-compatible API
 - **Video recognition requires ffmpeg** (install on server, then set `media.auto.video.enabled = true`)
+- Video recognition supports **light / full** modes (default: light)
 - Small text files can be previewed automatically
 
 ## Send queue & operation logs (optional)

package/README.md

@@ -95,10 +95,12 @@ openclaw gateway restart
 - `/sendfile`：发送服务器文件（支持多个绝对路径）
   - 支持目录：自动打包为 zip 后发送
   - 示例：`/sendfile /tmp/openclaw-wecom /home/shu/Desktop/report.pdf`
+  - 也支持自然语言：`把这个文件发给我 image-xxx.jpg`（默认仅在 `media.tempDir` 内匹配）
 
 ## 多媒体自动识别（可选）
 - **语音收发不需要 API**，只有开启“语音自动转写”才需要 OpenAI 兼容接口
 - **视频识别需要 ffmpeg**（服务器已安装后，将 `media.auto.video.enabled` 设为 `true`）
+- 视频识别支持 **light / full** 两种模式（默认 light）
 - 文本文件可自动预览（小文件直接读入）
 
 ## 发送队列与操作日志（可选）

package/README.zh.md

@@ -95,10 +95,12 @@ openclaw gateway restart
 - `/sendfile`：发送服务器文件（支持多个绝对路径）
   - 支持目录：自动打包为 zip 后发送
   - 示例：`/sendfile /tmp/openclaw-wecom /home/shu/Desktop/report.pdf`
+  - 也支持自然语言：`把这个文件发给我 image-xxx.jpg`（默认仅在 `media.tempDir` 内匹配）
 
 ## 多媒体自动识别（可选）
 - **语音收发不需要 API**，只有开启“语音自动转写”才需要 OpenAI 兼容接口
 - **视频识别需要 ffmpeg**（服务器已安装后，将 `media.auto.video.enabled` 设为 `true`）
+- 视频识别支持 **light / full** 两种模式（默认 light）
 - 文本文件可自动预览（小文件直接读入）
 
 ## 发送队列与操作日志（可选）

package/docs/INSTALL.md

@@ -131,6 +131,7 @@ openclaw gateway restart
 ### /sendfile（文件与文件夹）
 - `/sendfile` 仅支持 **服务器绝对路径**
 - 目录会自动打包为 zip 再发送
+- 自然语言也可触发：`把这个文件发给我 image-xxx.jpg`（默认仅在 `media.tempDir` 内匹配）
 
 示例：
 ```

package/docs/wecom.config.full.example.json

@@ -39,7 +39,12 @@
           "video": {
             "enabled": true,
             "ffmpegPath": "ffmpeg",
-            "maxBytes": 104857600
+            "maxBytes": 104857600,
+            "mode": "light",
+            "frames": 5,
+            "intervalSec": 2,
+            "maxDurationSec": 60,
+            "maxFrames": 30
           }
         }
       },

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@marshulll/openclaw-wecom",
-  "version": "0.1.22",
+  "version": "0.1.24",
   "type": "module",
   "description": "OpenClaw WeCom channel plugin (intelligent bot + internal app)",
   "author": "OpenClaw",

package/wecom/src/config-schema.ts

@@ -79,6 +79,12 @@ const accountSchema = z.object({
         enabled: z.boolean().optional(),
         ffmpegPath: z.string().optional(),
         maxBytes: z.number().optional(),
+        mode: z.enum(["light", "full"]).optional(),
+        frames: z.number().optional(),
+        intervalSec: z.number().optional(),
+        maxDurationSec: z.number().optional(),
+        maxFrames: z.number().optional(),
+        includeAudio: z.boolean().optional(),
       }).optional(),
     }).optional(),
   }).optional(),
@@ -155,6 +161,12 @@ export const WecomConfigSchema = ensureJsonSchema(z.object({
         enabled: z.boolean().optional(),
         ffmpegPath: z.string().optional(),
         maxBytes: z.number().optional(),
+        mode: z.enum(["light", "full"]).optional(),
+        frames: z.number().optional(),
+        intervalSec: z.number().optional(),
+        maxDurationSec: z.number().optional(),
+        maxFrames: z.number().optional(),
+        includeAudio: z.boolean().optional(),
       }).optional(),
     }).optional(),
   }).optional(),

package/wecom/src/media-auto.ts

@@ -1,4 +1,4 @@
-import { readFile, stat, writeFile, mkdtemp, rm } from "node:fs/promises";
+import { readFile, stat, writeFile, mkdtemp, rm, readdir } from "node:fs/promises";
 import { spawn } from "node:child_process";
 import { tmpdir } from "node:os";
 import { basename, extname, join } from "node:path";
@@ -49,6 +49,12 @@ export type ResolvedAutoFileConfig = {
 export type ResolvedAutoVideoConfig = {
   ffmpegPath: string;
   maxBytes?: number;
+  mode: "light" | "full";
+  frames: number;
+  intervalSec: number;
+  maxDurationSec: number;
+  maxFrames: number;
+  includeAudio: boolean;
 };
 
 export function resolveAutoAudioConfig(cfg: WecomAccountConfig): ResolvedAutoAudioConfig | null {
@@ -86,9 +92,29 @@ export function resolveAutoFileConfig(cfg: WecomAccountConfig): ResolvedAutoFile
 export function resolveAutoVideoConfig(cfg: WecomAccountConfig): ResolvedAutoVideoConfig | null {
   const video = cfg.media?.auto?.video;
   if (!cfg.media?.auto?.enabled || !video?.enabled) return null;
+  const mode = video.mode === "full" ? "full" : "light";
+  const maxDurationSec = typeof video.maxDurationSec === "number" && video.maxDurationSec > 0
+    ? video.maxDurationSec
+    : mode === "full" ? 120 : 60;
+  const frames = typeof video.frames === "number" && video.frames > 0
+    ? video.frames
+    : mode === "full" ? 12 : 5;
+  const intervalSec = typeof video.intervalSec === "number" && video.intervalSec > 0
+    ? video.intervalSec
+    : Math.max(1, Math.round(maxDurationSec / Math.max(frames, 1)));
+  const maxFrames = typeof video.maxFrames === "number" && video.maxFrames > 0
+    ? video.maxFrames
+    : mode === "full" ? 30 : frames;
+  const includeAudio = video.includeAudio === true;
   return {
     ffmpegPath: video.ffmpegPath?.trim() || "ffmpeg",
     maxBytes: typeof video.maxBytes === "number" && video.maxBytes > 0 ? video.maxBytes : undefined,
+    mode,
+    frames,
+    intervalSec,
+    maxDurationSec,
+    maxFrames,
+    includeAudio,
   };
 }
 
@@ -182,6 +208,39 @@ async function runFfmpegExtractFrame(params: {
   });
 }
 
+async function runFfmpegExtractFrames(params: {
+  ffmpegPath: string;
+  videoPath: string;
+  outputPattern: string;
+  fps: number;
+  maxDurationSec: number;
+}): Promise<void> {
+  await new Promise<void>((resolve, reject) => {
+    const proc = spawn(params.ffmpegPath, [
+      "-y",
+      "-i",
+      params.videoPath,
+      "-t",
+      String(params.maxDurationSec),
+      "-vf",
+      `fps=${params.fps}`,
+      "-q:v",
+      "2",
+      params.outputPattern,
+    ]);
+    proc.on("error", reject);
+    proc.on("close", (code) => {
+      if (code === 0) resolve();
+      else reject(new Error(`ffmpeg exited with code ${code ?? "unknown"}`));
+    });
+  });
+}
+
+function truncateText(text: string, maxChars: number): string {
+  if (text.length <= maxChars) return text;
+  return `${text.slice(0, maxChars)}…`;
+}
+
 export async function summarizeVideoWithVision(params: {
   cfg: ResolvedAutoVideoConfig;
   account: WecomAccountConfig;
@@ -195,19 +254,61 @@ export async function summarizeVideoWithVision(params: {
   const tempDir = await mkdtemp(join(tmpdir(), "openclaw-wecom-frame-"));
   const framePath = join(tempDir, `${basename(params.videoPath)}.jpg`);
   try {
-    await runFfmpegExtractFrame({
-      ffmpegPath: params.cfg.ffmpegPath,
-      videoPath: params.videoPath,
-      framePath,
-    });
-    const buffer = await readFile(framePath);
-    if (!buffer.length) return null;
-    const summary = await describeImageWithVision({
-      config: visionConfig,
-      buffer,
-      mimeType: "image/jpeg",
-    });
-    return summary ?? null;
+    const summaries: string[] = [];
+    if (params.cfg.mode === "light") {
+      const fps = Math.max(0.05, params.cfg.frames / Math.max(params.cfg.maxDurationSec, 1));
+      await runFfmpegExtractFrames({
+        ffmpegPath: params.cfg.ffmpegPath,
+        videoPath: params.videoPath,
+        outputPattern: join(tempDir, "frame-%03d.jpg"),
+        fps,
+        maxDurationSec: params.cfg.maxDurationSec,
+      });
+      const frames = (await readdir(tempDir))
+        .filter((name) => name.startsWith("frame-") && name.endsWith(".jpg"))
+        .sort()
+        .slice(0, params.cfg.maxFrames);
+      for (const frame of frames) {
+        const buffer = await readFile(join(tempDir, frame));
+        if (!buffer.length) continue;
+        const summary = await describeImageWithVision({
+          config: visionConfig,
+          buffer,
+          mimeType: "image/jpeg",
+        });
+        if (summary) summaries.push(summary);
+      }
+    } else {
+      const fps = Math.max(0.1, 1 / Math.max(params.cfg.intervalSec, 1));
+      await runFfmpegExtractFrames({
+        ffmpegPath: params.cfg.ffmpegPath,
+        videoPath: params.videoPath,
+        outputPattern: join(tempDir, "frame-%03d.jpg"),
+        fps,
+        maxDurationSec: params.cfg.maxDurationSec,
+      });
+      const frames = (await readdir(tempDir))
+        .filter((name) => name.startsWith("frame-") && name.endsWith(".jpg"))
+        .sort()
+        .slice(0, params.cfg.maxFrames);
+      for (const frame of frames) {
+        const buffer = await readFile(join(tempDir, frame));
+        if (!buffer.length) continue;
+        const summary = await describeImageWithVision({
+          config: visionConfig,
+          buffer,
+          mimeType: "image/jpeg",
+        });
+        if (summary) summaries.push(summary);
+      }
+    }
+
+    if (summaries.length === 0) return null;
+    const unique = Array.from(new Set(summaries.map((s) => s.trim()).filter(Boolean)));
+    const maxChars = 1600;
+    const lines = unique.slice(0, params.cfg.maxFrames).map((s, idx) => `${idx + 1}. ${s}`);
+    const joined = truncateText(lines.join("\n"), maxChars);
+    return `关键帧概述（${unique.length}帧）\n${joined}`;
   } catch {
     return null;
   } finally {

package/wecom/src/types.ts

@@ -79,6 +79,12 @@ export type WecomAccountConfig = {
         enabled?: boolean;
         ffmpegPath?: string;
         maxBytes?: number;
+        mode?: "light" | "full";
+        frames?: number;
+        intervalSec?: number;
+        maxDurationSec?: number;
+        maxFrames?: number;
+        includeAudio?: boolean;
       };
     };
   };

package/wecom/src/wecom-app.ts

@@ -130,6 +130,139 @@ function sleep(ms: number): Promise<void> {
   return new Promise((resolve) => setTimeout(resolve, ms));
 }
 
+function resolveSendIntervalMs(target: WecomWebhookTarget): number {
+  const interval = target.account.config.sendQueue?.intervalMs;
+  return typeof interval === "number" && interval >= 0 ? interval : 400;
+}
+
+function extractFilenameCandidates(text: string): string[] {
+  const candidates = new Set<string>();
+  const normalized = text.replace(/[，,；;|]/g, " ");
+  const regex = /(?:\/|file:\/\/)?[A-Za-z0-9._-]+\.[A-Za-z0-9]{1,8}/g;
+  for (const match of normalized.matchAll(regex)) {
+    const value = match[0];
+    if (value) candidates.add(value.replace(/^file:\/\//, ""));
+  }
+  return Array.from(candidates);
+}
+
+async function tryHandleNaturalFileSend(params: {
+  target: WecomWebhookTarget;
+  text: string;
+  fromUser: string;
+  chatId?: string;
+  isGroup: boolean;
+}): Promise<boolean> {
+  const { target, text, fromUser, chatId, isGroup } = params;
+  if (!text || text.trim().startsWith("/")) return false;
+  if (!/(发给我|发送给我|发我|给我)/.test(text)) return false;
+  const names = extractFilenameCandidates(text);
+  if (names.length === 0) return false;
+
+  const tempDir = resolveMediaTempDir(target);
+  let dirEntries: string[] = [];
+  try {
+    dirEntries = await readdir(tempDir);
+  } catch {
+    dirEntries = [];
+  }
+  const dirSet = new Set(dirEntries);
+
+  const resolved: string[] = [];
+  const missing: string[] = [];
+  for (const name of names) {
+    let fullPath = "";
+    if (name.startsWith("/")) {
+      fullPath = name;
+    } else if (dirSet.has(name)) {
+      fullPath = join(tempDir, name);
+    }
+    if (!fullPath) {
+      missing.push(name);
+      continue;
+    }
+    try {
+      const info = await stat(fullPath);
+      if (info.isFile()) {
+        resolved.push(fullPath);
+      } else {
+        missing.push(name);
+      }
+    } catch {
+      missing.push(name);
+    }
+  }
+
+  if (resolved.length === 0) {
+    const hint = dirEntries.length ? `可用文件示例：${dirEntries.slice(0, 5).join(", ")}` : "当前目录无可用文件";
+    await sendWecomText({
+      account: target.account,
+      toUser: fromUser,
+      chatId: isGroup ? chatId : undefined,
+      text: `未找到指定文件：${missing.join(", ")}。\n${hint}`,
+    });
+    return true;
+  }
+
+  const maxBytes = resolveMediaMaxBytes(target);
+  const intervalMs = resolveSendIntervalMs(target);
+  let sent = 0;
+  const failed: string[] = [];
+
+  for (const path of resolved) {
+    try {
+      const info = await stat(path);
+      if (maxBytes && info.size > maxBytes) {
+        failed.push(`${basename(path)}(过大)`);
+        continue;
+      }
+      const buffer = await readFile(path);
+      const filename = basename(path) || "file.bin";
+      const mediaId = await uploadWecomMedia({
+        account: target.account,
+        type: "file",
+        buffer,
+        filename,
+      });
+      await sendWecomFile({
+        account: target.account,
+        toUser: fromUser,
+        chatId: isGroup ? chatId : undefined,
+        mediaId,
+      });
+      sent += 1;
+      await appendOperationLog(target, {
+        action: "natural-sendfile",
+        accountId: target.account.accountId,
+        toUser: fromUser,
+        chatId,
+        path,
+        size: info.size,
+      });
+      if (intervalMs) await sleep(intervalMs);
+    } catch (err) {
+      failed.push(basename(path));
+      await appendOperationLog(target, {
+        action: "natural-sendfile",
+        accountId: target.account.accountId,
+        toUser: fromUser,
+        chatId,
+        path,
+        error: String(err),
+      });
+    }
+  }
+
+  const summary = `已发送 ${sent} 个文件${failed.length ? `，失败：${failed.join(", ")}` : ""}${missing.length ? `，未找到：${missing.join(", ")}` : ""}`;
+  await sendWecomText({
+    account: target.account,
+    toUser: fromUser,
+    chatId: isGroup ? chatId : undefined,
+    text: summary,
+  });
+  return true;
+}
+
 async function appendOperationLog(target: WecomWebhookTarget, entry: Record<string, unknown>): Promise<void> {
   const logPath = target.account.config.operations?.logPath?.trim();
   if (!logPath) return;
@@ -861,6 +994,17 @@ async function processAppMessage(params: {
     if (handled) return;
   }
 
+  if (msgType === "text") {
+    const handled = await tryHandleNaturalFileSend({
+      target,
+      text: messageText,
+      fromUser,
+      chatId,
+      isGroup,
+    });
+    if (handled) return;
+  }
+
   try {
     await startAgentForApp({
       target,