Skip to main content
Glama
al-one

MCP Server for notify to weixin / telegram / bark / lark

wework_app_send_file

Send files through WeChat Work application notifications to specified users or entire teams using file URLs.

Instructions

通过企业微信应用号发送发送文件消息

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
urlYes文件URL
touserNo接收消息的成员ID,多个用`|`分隔,为`@all`时向该企业应用全部成员发送,默认从环境变量获取

Implementation Reference

  • The handler function for the 'wework_app_send_file' MCP tool. It defines the input schema using Pydantic Field for the file URL and target user, and executes by calling the shared wework_send_media helper with type 'file'.
    @mcp.tool(
        title="企业微信应用号-发送文件消息",
        description="通过企业微信应用号发送发送文件消息",
    )
    def wework_app_send_file(
        url: str = Field(description="文件URL"),
        touser: str = FIELD_TO_USER,
    ):
        return wework_send_media(touser, url, "file")
  • Helper function that implements the core logic for sending media (image, video, voice, file) via WeWork app: downloads from URL, determines type, uploads to media API using access token, then sends the media_id to the specified user.
    def wework_send_media(touser, url: str, msgtype=None):
        if msgtype:
            pass
        elif '.jpg' in url.lower() or '.jpeg' in url.lower() or '.png' in url.lower():
            msgtype = 'image'
        elif '.mp4' in url.lower():
            msgtype = 'video'
        elif '.arm' in url.lower():
            msgtype = 'voice'
        else:
            msgtype = 'file'
        res = requests.get(url, timeout=120)
        res.raise_for_status()
        file = io.BytesIO(res.content)
        mine = res.headers.get("content-type") or "application/octet-stream"
        res = requests.post(
            f"{WEWORK_BASE_URL}/cgi-bin/media/upload",
            params={"type": msgtype, "access_token": get_access_token()},
            files={"media": ("filename", file, mine)},
            timeout=120,
        )
        media = res.json() or {}
        if not (media_id := media.get("media_id")):
            return media
        res = requests.post(
            f"{WEWORK_BASE_URL}/cgi-bin/message/send?access_token={get_access_token()}",
            json={
                "touser": touser or WEWORK_APP_TOUSER,
                "agentid": WEWORK_APP_AGENTID,
                "msgtype": msgtype,
                msgtype: {"media_id": media_id},
            },
        )
        return res.json()
  • Cached helper function to obtain the WeWork application access token, required for API calls like media upload and message sending.
    @cached(TTLCache(maxsize=1, ttl=3600))
    def get_access_token():
        res = requests.get(
            f"{WEWORK_BASE_URL}/cgi-bin/gettoken",
            params={"corpid": WEWORK_APP_CORPID, "corpsecret": WEWORK_APP_SECRET},
            timeout=60,
        )
        return res.json().get("access_token")
  • Calls add_tools from the wework module to register all WeWork-related tools, including 'wework_app_send_file', onto the FastMCP instance.
    wework.add_tools(mcp)

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/al-one/mcp-notify'

If you have feedback or need assistance with the MCP directory API, please join our Discord server