feat(monitor): 抖音 Web 接口客户端 —— 绕开爬虫子进程,直接发 HTTP
移植自 mac-agent-os 的 mediacrawler_adapter:不起子进程、不开页面,用浏览器里那份
登录态直接调抖音 Web 接口。产物键名照抄 store/douyin,所以 ingest 那条链路一个字不用改。
**目前能用的(真环境实测,非推断)**:
profile/other : 200, 7075 字节 —— 博主主页指标(粉丝/获赞/作品数/昵称)
aweme/detail : 200, 45425 字节 —— 单条作品详情(含点赞/评论/收藏/分享)
**目前不能用的:作品列表 `aweme/post`。** 两个互相独立的原因:
1. 这个接口被抖音单独升级成了真校验:不带 x-tt-argus 回 403「Uifid Not Found」,
带上 dummy 值回 200 + **空 body**。也就是说「头在不在」骗得过,「真校验」过不了。
同一套头打 profile/other 和 aweme/detail 都是通的 —— 抖音是挑着接口加保护的,
挑中的恰好是「批量拉作品列表」这个最敏感的动作。
2. 改走页面截获也不行:CDP 浏览器打开博主主页会落到「验证码中间页」(当天大量探测的
代价,过几小时要重测)。
所以现在的边界是:**已知作品的指标刷新能做,自动发现新作品做不了**。
**排查中控住变量后得到的两条事实**(都写进注释了):
· `Accept` / `Accept-Language` / `Referer` 才是主页接口能返回真数据的原因 —— 只有
UA+client hints+Cookie 时是 200 但仅 121 字节的空壳,补上这三个头变 7074 字节。
(我先前猜的 sec-ch-ua 不是关键。)
· 因此 UA 与 client hints 必须**成套地取自同一个浏览器**,所以 BrowserIdentity 一次
从 CDP 取齐 cookie + UA + hints,而不是各自写死。
「200 + 空 body 必须当场报错」也是刻意写死的:放过去它会在下游变成「这个博主没作品」,
把一次失败伪装成一条正常结果 —— 爬虫那条路正是这么栽的,还被翻译成「账号被封」。
测试 +11:cookie 解析、请求头成套性(含 uifid 缺失/回退)、产物键名与 store 对齐、
以及 _get 的三条失败路径(空 body / 403 带网关原话 / 正常返回)。
This commit is contained in:
@@ -0,0 +1,177 @@
|
||||
"""
|
||||
agentos tool — 工具管理
|
||||
|
||||
子命令: list, trae
|
||||
"""
|
||||
|
||||
import os
|
||||
import subprocess
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
from .utils import get_sync_root, info, ok, warn, err, banner
|
||||
|
||||
|
||||
TOOL_CATEGORIES = {
|
||||
"00_setup": "安装部署",
|
||||
"01_system": "系统诊断与健康检查",
|
||||
"02_browser": "浏览器自动化",
|
||||
"03_ocr": "OCR 识别",
|
||||
"04_media": "媒体处理",
|
||||
"05_crawl": "网页抓取与爬虫",
|
||||
"06_mobile": "移动端工具",
|
||||
"07_matrix": "矩阵养号",
|
||||
"08_trae_agent": "Trae AI 编程助手",
|
||||
}
|
||||
|
||||
|
||||
def _get_trae_script() -> Path:
|
||||
"""查找 trae_agent.sh 脚本路径"""
|
||||
sync_root = get_sync_root()
|
||||
candidates = [
|
||||
sync_root / "05_tools" / "08_trae_agent" / "trae_agent.sh",
|
||||
sync_root / "05_tools" / "08_trae_agent" / "trae_agent.sh",
|
||||
]
|
||||
for p in candidates:
|
||||
if p.exists():
|
||||
return p
|
||||
# Fallback: 直接使用 agent-local 路径
|
||||
fallback = Path.home() / "workbuddy-agent-os" / "agent-local" / "tools" / "trae-agent"
|
||||
script = fallback.parent.parent / "agent-sync" / "05_tools" / "08_trae_agent" / "trae_agent.sh"
|
||||
if script.exists():
|
||||
return script
|
||||
return candidates[0]
|
||||
|
||||
|
||||
def cmd_list():
|
||||
"""列出所有可用工具"""
|
||||
banner()
|
||||
tools_root = get_sync_root() / "05_tools"
|
||||
|
||||
if not tools_root.exists():
|
||||
warn(f"工具目录不存在: {tools_root}")
|
||||
return
|
||||
|
||||
total = 0
|
||||
for category_dir in sorted(tools_root.iterdir()):
|
||||
if not category_dir.is_dir() or category_dir.name.startswith("."):
|
||||
continue
|
||||
|
||||
cat_name = TOOL_CATEGORIES.get(category_dir.name, category_dir.name)
|
||||
scripts = sorted(category_dir.rglob("*.py")) + sorted(category_dir.rglob("*.sh"))
|
||||
|
||||
# 过滤掉 __pycache__ 和 __init__.py
|
||||
scripts = [s for s in scripts if "__pycache__" not in s.parts and s.name != "__init__.py"]
|
||||
|
||||
if not scripts:
|
||||
continue
|
||||
|
||||
print(f"\n{cat_name} ({category_dir.name}/):")
|
||||
for s in scripts:
|
||||
rel = s.relative_to(tools_root)
|
||||
# 显示简要描述(读取文件第一行注释)
|
||||
desc = ""
|
||||
try:
|
||||
with open(s) as f:
|
||||
first_line = f.readline().strip()
|
||||
if first_line.startswith("#"):
|
||||
desc = first_line.lstrip("# ").strip()
|
||||
elif first_line.startswith('"""'):
|
||||
desc = s.read_text().split('"""')[1].strip().split("\n")[0].strip()
|
||||
except Exception:
|
||||
pass
|
||||
desc = desc[:50] if desc else ""
|
||||
print(f" 📄 {rel}")
|
||||
if desc:
|
||||
print(f" {desc}")
|
||||
total += 1
|
||||
|
||||
print(f"\n共 {total} 个工具脚本")
|
||||
print()
|
||||
|
||||
|
||||
def cmd_trae_install():
|
||||
"""安装 trae-agent"""
|
||||
sync_root = get_sync_root()
|
||||
install_script = sync_root / "05_tools" / "08_trae_agent" / "install_trae_agent.sh"
|
||||
if not install_script.exists():
|
||||
err(f"安装脚本不存在: {install_script}")
|
||||
return
|
||||
info("正在安装 trae-agent...")
|
||||
result = subprocess.run(["bash", str(install_script)], cwd=str(install_script.parent))
|
||||
if result.returncode == 0:
|
||||
ok("trae-agent 安装完成")
|
||||
else:
|
||||
err("安装失败")
|
||||
|
||||
|
||||
def cmd_trae_run(args):
|
||||
"""执行 trae-agent 任务"""
|
||||
script = _get_trae_script()
|
||||
if not script.exists():
|
||||
err(f"trae_agent.sh 不存在: {script}")
|
||||
err("请先运行: agentos tool trae install")
|
||||
return
|
||||
|
||||
# 绕过系统代理(macOS 系统代理端口 6478 会拦截本地 LLM 请求)
|
||||
env = {**subprocess._clean_environ() if hasattr(subprocess, '_clean_environ') else {}, **os.environ}
|
||||
env["NO_PROXY"] = "localhost,127.0.0.1,::1"
|
||||
env["no_proxy"] = "localhost,127.0.0.1,::1"
|
||||
|
||||
cmd = ["bash", str(script), "run"]
|
||||
if args.provider:
|
||||
cmd.extend(["-p", args.provider])
|
||||
if args.model:
|
||||
cmd.extend(["-m", args.model])
|
||||
if args.file:
|
||||
cmd.extend(["-f", args.file])
|
||||
if args.task:
|
||||
cmd.append(" ".join(args.task))
|
||||
|
||||
result = subprocess.run(cmd, env=env)
|
||||
sys.exit(result.returncode)
|
||||
|
||||
|
||||
def cmd_trae_interactive():
|
||||
"""启动 trae-agent 交互模式"""
|
||||
script = _get_trae_script()
|
||||
if not script.exists():
|
||||
err(f"trae_agent.sh 不存在: {script}")
|
||||
err("请先运行: agentos tool trae install")
|
||||
return
|
||||
subprocess.run(["bash", str(script), "interactive"])
|
||||
|
||||
|
||||
def cmd_trae_config():
|
||||
"""查看 trae-agent 配置"""
|
||||
script = _get_trae_script()
|
||||
if not script.exists():
|
||||
err("trae-agent 未安装")
|
||||
return
|
||||
subprocess.run(["bash", str(script), "show-config"])
|
||||
|
||||
|
||||
def run(args):
|
||||
cmd = args.tool_cmd
|
||||
if not cmd:
|
||||
banner()
|
||||
print("agentos tool 子命令: list, trae")
|
||||
print()
|
||||
return
|
||||
if cmd == "list":
|
||||
cmd_list()
|
||||
elif cmd == "trae":
|
||||
trae_cmd = getattr(args, "trae_cmd", None)
|
||||
if not trae_cmd:
|
||||
banner()
|
||||
print("agentos tool trae 子命令: run, interactive, config, install")
|
||||
print()
|
||||
return
|
||||
if trae_cmd == "run":
|
||||
cmd_trae_run(args)
|
||||
elif trae_cmd == "interactive":
|
||||
cmd_trae_interactive()
|
||||
elif trae_cmd == "config":
|
||||
cmd_trae_config()
|
||||
elif trae_cmd == "install":
|
||||
cmd_trae_install()
|
||||
Reference in New Issue
Block a user