我有个想法，不接入各种AI，自定义回复内容

nomoneynolife commented 9 months ago

⚠️ 搜索是否存在类似issue

[X] 我已经搜索过issues和disscussions，没有发现相似issue

总结

有些知识AI肯定没有，对吧

举例

比如，通过企业微信自建接入后：用户问，大佬，年假是怎么计算的啊。，，，这时AI就不行了（每个公司制度不一样）我直接自定义回复内容，年假的计算方式为：年休假天数=（当年已在职天数/365天）*当年度年休假总天数跪求大佬们开发，自定义功能，

动机

有实际需求

sealot commented 9 months ago

认同需求。思路或建议：是否可以对接自己的API，给出对接参数和调用标准。我自己开发API可以先将用户信息做处理，例如：识别用户、鉴别问题，或者做langchain应用查询知识库，再自己调用GPT或Qwen接口，将结果按标准传回

huluohu commented 9 months ago

fastgpt，你值得拥有

kokojacket commented 9 months ago

接入fastgpt知识库就好了

nomoneynolife commented 9 months ago

接入fastgpt知识库就好了

收费的哇

kokojacket commented 9 months ago

接入fastgpt知识库就好了

收费的哇

你自己部署啊

ju123123 commented 8 months ago

不是有个自带的keyword插件吗，自己改造下，加个功能可以增删改查

    def on_handle_context(self, e_context: EventContext):
        if e_context["context"].type != ContextType.TEXT:
            return
        content = e_context["context"].content.strip()
        logger.debug("[keyword] on_handle_context. content: %s" % content)
        if str(co    def on_handle_context(self, e_context: EventContext):
        if e_context["context"].type != ContextType.TEXT:
            return
        content = e_context["context"].content.strip()
        logger.debug("[keyword] on_handle_context. content: %s" % content)
        if str(content).startswith("delkeyword"):
            logger.debug("delkeyword============================================")
            content_str = str(content)
            key = content_str[len("delkeyword "):]
            # 读取config.json文件
            with open("/app/plugins/keyword/config.json", "r") as f:
                jsonconfig = json.load(f)
                logger.debug(jsonconfig)
            #添加关键字和值到config中的keyword数组
            if key in jsonconfig["keyword"]:
                del jsonconfig["keyword"][key]
            self.keyword = jsonconfig["keyword"]
            # 写回config.json文件
            with open("/app/plugins/keyword/config.json", "w", encoding='utf-8') as f:
                json.dump(jsonconfig, f, indent=4, ensure_ascii=False)
            reply = Reply()
            reply.type = ReplyType.TEXT
            reply.content = f"关键字 '{key}' 已删除'。"
            e_context["reply"] = reply
            e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑
        if str(content).startswith("setkeyword"):
            logger.debug("setkeyword============================================")
            content_str = str(content)
            key_value_pair = content_str[len("setkeyword "):].split(":::")
            logger.debug("[keyword] {}".format(key_value_pair))
            if len(key_value_pair) == 2:
                key = key_value_pair[0].strip()
                value = key_value_pair[1].strip()
                logger.debug("[key] {}".format(key))
                logger.debug("[value] {}".format(value))
                # 读取config.json文件
                with open("/app/plugins/keyword/config.json", "r") as f:
                    jsonconfig = json.load(f)
                    logger.debug(jsonconfig)
                #添加关键字和值到config中的keyword数组
                jsonconfig["keyword"][key] = value
                self.keyword = jsonconfig["keyword"]
                # 写回config.json文件
                with open("/app/plugins/keyword/config.json", "w", encoding='utf-8') as f:
                    json.dump(jsonconfig, f, indent=4, ensure_ascii=False)
                reply = Reply()
                reply.type = ReplyType.TEXT
                reply.content = f"关键字 '{key}' 已成功设置为 '{value}'。"
                e_context["reply"] = reply
                e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑
        if content in self.keyword:
            logger.info(f"[keyword] 匹配到关键字【{content}】")
            reply_text = self.keyword[content]

            # 判断匹配内容的类型
            if (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".jpg", ".jpeg", ".png", ".img", ".gif"]):
            # 如果是以 http:// 或 https:// 开头，且".jpg", ".jpeg", ".png", ".gif", ".img"结尾，则认为是图片 URL。
                reply = Reply()
                reply.type = ReplyType.IMAGE_URL
                reply.content = reply_text

            elif (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".pdf", ".doc", ".docx", ".xls", "xlsx",".zip", ".rar"]):
            # 如果是以 http:// 或 https:// 开头，且".pdf", ".doc", ".docx", ".xls", "xlsx",".zip", ".rar"结尾，则下载文件到tmp目录并发送给用户
                file_path = "tmp"
                if not os.path.exists(file_path):
                    os.makedirs(file_path)
                file_name = reply_text.split("/")[-1]  # 获取文件名
                file_path = os.path.join(file_path, file_name)
                response = requests.get(reply_text)
                with open(file_path, "wb") as f:
                    f.write(response.content)
                #channel/wechat/wechat_channel.py和channel/wechat_channel.py中缺少ReplyType.FILE类型。
                reply = Reply()
                reply.type = ReplyType.FILE
                reply.content = file_path

            elif (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".mp4"]):
            # 如果是以 http:// 或 https:// 开头，且".mp4"结尾，则下载视频到tmp目录并发送给用户
                reply = Reply()
                reply.type = ReplyType.VIDEO_URL
                reply.content = reply_text

            else:
            # 否则认为是普通文本
                reply = Reply()
                reply.type = ReplyType.TEXT
                reply.content = reply_text

            e_context["reply"] = reply
            e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑ntent).startswith("delkeyword"):
            logger.debug("delkeyword============================================")
            content_str = str(content)
            key = content_str[len("delkeyword "):]
            # 读取config.json文件
            with open("/app/plugins/keyword/config.json", "r") as f:
                jsonconfig = json.load(f)
                logger.debug(jsonconfig)
            #添加关键字和值到config中的keyword数组
            if key in jsonconfig["keyword"]:
                del jsonconfig["keyword"][key]
            self.keyword = jsonconfig["keyword"]
            # 写回config.json文件
            with open("/app/plugins/keyword/config.json", "w", encoding='utf-8') as f:
                json.dump(jsonconfig, f, indent=4, ensure_ascii=False)
            reply = Reply()
            reply.type = ReplyType.TEXT
            reply.content = f"关键字 '{key}' 已删除'。"
            e_context["reply"] = reply
            e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑
        if str(content).startswith("setkeyword"):
            logger.debug("setkeyword============================================")
            content_str = str(content)
            key_value_pair = content_str[len("setkeyword "):].split(":::")
            logger.debug("[keyword] {}".format(key_value_pair))
            if len(key_value_pair) == 2:
                key = key_value_pair[0].strip()
                value = key_value_pair[1].strip()
                logger.debug("[key] {}".format(key))
                logger.debug("[value] {}".format(value))
                # 读取config.json文件
                with open("/app/plugins/keyword/config.json", "r") as f:
                    jsonconfig = json.load(f)
                    logger.debug(jsonconfig)
                #添加关键字和值到config中的keyword数组
                jsonconfig["keyword"][key] = value
                self.keyword = jsonconfig["keyword"]
                # 写回config.json文件
                with open("/app/plugins/keyword/config.json", "w", encoding='utf-8') as f:
                    json.dump(jsonconfig, f, indent=4, ensure_ascii=False)
                reply = Reply()
                reply.type = ReplyType.TEXT
                reply.content = f"关键字 '{key}' 已成功设置为 '{value}'。"
                e_context["reply"] = reply
                e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑
        if content in self.keyword:
            logger.info(f"[keyword] 匹配到关键字【{content}】")
            reply_text = self.keyword[content]

            # 判断匹配内容的类型
            if (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".jpg", ".jpeg", ".png", ".img", ".gif"]):
            # 如果是以 http:// 或 https:// 开头，且".jpg", ".jpeg", ".png", ".gif", ".img"结尾，则认为是图片 URL。
                reply = Reply()
                reply.type = ReplyType.IMAGE_URL
                reply.content = reply_text

            elif (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".pdf", ".doc", ".docx", ".xls", "xlsx",".zip", ".rar"]):
            # 如果是以 http:// 或 https:// 开头，且".pdf", ".doc", ".docx", ".xls", "xlsx",".zip", ".rar"结尾，则下载文件到tmp目录并发送给用户
                file_path = "tmp"
                if not os.path.exists(file_path):
                    os.makedirs(file_path)
                file_name = reply_text.split("/")[-1]  # 获取文件名
                file_path = os.path.join(file_path, file_name)
                response = requests.get(reply_text)
                with open(file_path, "wb") as f:
                    f.write(response.content)
                #channel/wechat/wechat_channel.py和channel/wechat_channel.py中缺少ReplyType.FILE类型。
                reply = Reply()
                reply.type = ReplyType.FILE
                reply.content = file_path

            elif (reply_text.startswith("http://") or reply_text.startswith("https://")) and any(reply_text.endswith(ext) for ext in [".mp4"]):
            # 如果是以 http:// 或 https:// 开头，且".mp4"结尾，则下载视频到tmp目录并发送给用户
                reply = Reply()
                reply.type = ReplyType.VIDEO_URL
                reply.content = reply_text

            else:
            # 否则认为是普通文本
                reply = Reply()
                reply.type = ReplyType.TEXT
                reply.content = reply_text

            e_context["reply"] = reply
            e_context.action = EventAction.BREAK_PASS  # 事件结束，并跳过处理context的默认逻辑

AlberHu commented 6 months ago

mark一下

Z-XINY commented 6 months ago

你可以仿照bot文件夹里的模型调用方式，自己实现api接口，简单点直接通过http请求访问参考baidu_unit_bot.py的实现方式，在后端你可以部署自己的AI模型，配好请求接口就能用了。我这边试过本地的botpress模型通过请求调用然后发送内容到公众号上，思路是可行的。

zhayujie / chatgpt-on-wechat

我有个想法，不接入各种AI，自定义回复内容 #1690

⚠️ 搜索是否存在类似issue

总结

举例

动机