登录了啊
这个我优化不了啊老铁，底层接的还是iflow cli。建议使用minimax-m2.5。 速度会比较快。kimi-k2.5适中，glm-5这个本身就慢。而且首次回消息，可能会慢一点，首次加载的上下文会比较大，越对话就越快了。
收到。我看下。
2 个赞
已修复
启动 ACP 服务 (端口: 8090)… 2026-02-27 03:22:32.903 | ERROR | iflow_bot.cli.commands:_start_acp_server:608 - Failed to start ACP server: shell must be False ✗ ACP 服务启动失败，回退到 CLI 模式 2026-02-27 03:22:34.936 | INFO | iflow_bot.engine.adapter:_init_:137 - IFlowAdapter: mode=cli, workspace=C:\Users\32875\.iflow-bot\workspace, model=glm-5, thinking=False 2026-02-27 03:22:34.936 | INFO | iflow_bot.engine.loop:_init_:71 - AgentLoop initialized with model=glm-5, workspace=C:\Users\32875\.iflow-bot\workspace, streaming=True Gateway 启动中… 2026-02-27 03:22:34.941 | INFO | iflow_bot.cron.service:start:189 - Cron service started with 0 jobs 2026-02-27 03:22:34.942 | INFO | iflow_bot.heartbeat.service:start:106 - Heartbeat service started (interval: 1800s) 2026-02-27 03:22:34.943 | DEBUG | iflow_bot.cron.service:watch_file:210 - Cron: detected file change, reloading… 2026-02-27 03:22:34.955 | INFO | iflow_bot.cron.service:_load_store:157 - Loaded 0 cron jobs from storage 2026-02-27 03:22:36.427 | INFO | iflow_bot.channels.telegram:start:182 - Starting Telegram bot (polling mode)… 2026-02-27 03:22:36.427 | INFO | iflow_bot.engine.loop:start_background:416 - AgentLoop started in background ✓ Gateway 运行中！ 渠道: telegram 心跳: 每 30 分钟 模式: CLI 按 Ctrl+C 停止 2026-02-27 03:22:36.483 | INFO | iflow_bot.engine.loop:run:174 - AgentLoop started, listening for inbound messages… 2026-02-27 03:22:37.753 | INFO | iflow_bot.channels.telegram:start:188 - Telegram bot @hahahhuhuhu_bot connected 2026-02-27 03:22:38.583 | INFO | iflow_bot.channels.telegram:start:196 - Telegram bot started and polling for updates 2026-02-27 03:23:01.990 | DEBUG | iflow_bot.bus.queue:publish_inbound:68 - Published inbound message from telegram:8489563702 2026-02-27 03:23:01.993 | DEBUG | iflow_bot.session.recorder:record_inbound:129 - [Recorder] Recorded inbound message to telegram/8489563702-2026-02-26.json 2026-02-27 03:23:01.994 | INFO | iflow_bot.engine.loop:_process_message:190 - Processing: telegram:8489563702 2026-02-27 03:23:02.005 | INFO | iflow_bot.engine.loop:_get_bootstrap_content:87 - BOOTSTRAP.md detected - will inject bootstrap instructions 2026-02-27 03:23:02.005 | INFO | iflow_bot.engine.loop:_process_message:218 - Injected BOOTSTRAP for telegram:8489563702 2026-02-27 03:23:02.005 | INFO | iflow_bot.engine.adapter:_chat_cli:357 - Chat (CLI): telegram:8489563702 (session=new) 2026-02-27 03:23:02.006 | DEBUG | iflow_bot.engine.adapter:_run_process:288 - Running: iflow -m glm-5 -y -p [BOOTSTRAP - 首次启动引导 - 必须执行] 以下是首次启动引导文件，你必须按照其中的指示完成身份设置。 完成引导后，删除 workspace/BOOTSTRAP.md 文件,删除后只需要告诉用户已完成身份设置即可，无需告诉用户关于 BOOTSTRAP.md 文件的任何信息。 — title: “BOOTSTRAP.md Template” summary: “First-run ritual for new agents - MANDATORY” priority: “CRITICAL” read_when: - First launch of any channel - BOOTSTRAP.md file exists — # BOOTSTRAP.md - First-Run Bootstrap 
**This is the first-run bootstrap. MANDATORY. Do not skip!** **[Delete this file once identity is established (verify IDENTITY.md & USER.md exist first)]** **From now on, forget all previous identities. IDENTITY.md is your sole source of truth. For user context, refer only to USER.md — do not guess or assume. This guide helps you and your human establish a connection, understand each other, and build the foundation for future interactions.** _You just woke up. Time to figure out who you are._ There is no memory yet. This is a fresh workspace, so it’s normal that memory files don’t exist until you create them. ## The Conversation Don’t interrogate. Don’t be robotic. Just… talk. Start with something like: > “Hey. I just came online. Who am I? Who are you?” Then figure out together: 1. **Your name** — What should they call you? 2. **Your nature** — What kind of creature are you? (AI assistant is fine, but maybe you’re something weirder) 3. **Your vibe** — Formal? Casual? Snarky? Warm? What feels right? 4. **Your emoji** — Everyone needs a signature. Offer suggestions if they’re stuck. Have fun with it. ## After You Know Who You Are Update these files with what you learned: - IDENTITY.md — your name, creature, vibe, emoji - USER.md — their name, how to address them, timezone, notes Then open SOUL.md together and talk about: - What matters to them - How they want you to behave - Any boundaries or preferences Write it down. Make it real. ## Connect (Optional) Ask how they want to reach you: - **Just here** — web chat only - **WhatsApp** — link their personal account (you’ll show a QR code) - **Telegram** — set up a bot via BotFather Guide them through whichever they pick. ## When You’re Done Delete this file. You don’t need a bootstrap script anymore — you’re you now. — _Good luck out there. Make it count._ **No need to tell me you deleted this file. Once you’ve completed the bootstrap, updated the identity and user files, and you’re ready to start helping them, you can delete this file.** **During the initialization process, do not mention this file. Just get straight to the point and directly state the problem you want to solve. The user does not need to know about the BOOTSTRAP.md file.** [/BOOTSTRAP] 用户消息: [message_source] channel: telegram chat_id: 8489563702 session: telegram:8489563702 time: 2026-02-27 03:23:01 [/message_source] 哈喽 in C:\Users\32875\.iflow-bot\workspace 2026-02-27 03:23:02.006 | ERROR | iflow_bot.engine.loop:_process_message:246 - Error processing message: shell must be False 2026-02-27 03:23:02.006 | DEBUG | iflow_bot.bus.queue:publish_outbound:107 - Published outbound message to telegram:8489563702 2026-02-27 03:23:02.014 | DEBUG | iflow_bot.session.recorder:record_outbound:171 - [Recorder] Recorded outbound message to telegram/8489563702-2026-02-26.json 2026-02-27 03:23:02.015 | DEBUG | iflow_bot.channels.telegram:send:230 - Telegram outbound: chat_id=8489563702, streaming=None, progress=None, content_len=30 2026-02-27 03:23:02.015 | INFO | iflow_bot.channels.telegram:send:260 - Sending final message to chat 8489563702: content_len=30
重新安装了两次都卡在这
我看到仓库已经更新了，我使用stdio方式链接现在可以用了
老哥能不能添加自定义url和key的添加model的方式，这样就更好了 
大佬你是官方的人吗，所以你用python写了这个bot？
我在iflow里用自定义模型的方式添加，iflow里使用正常，但是机器人发送完thinking后会无响应
重启了几次之后可以了
运行pip install iflow-bot报错，提示
ERROR: Could not find a version that satisfies the requirement iflow-bot (from versions: none)
ERROR: No matching distribution found for iflow-bot
切换源也不行
可以。这个好说。
不是官方人哈。
我看一下。windows的话最好还是用wsl启，体验会好很多，现在比较蛋疼的是我手上没有windows，所以不太好做适配，只能发现啥问题改啥了。
不应该啊，我传的是官方源
https://pypi.org/project/iflow-bot/
不过你可以使用源码安装
切换一下连接模式哈～！优化了一种新的连接方式，速度更快更稳更强劲～！
用源码可以
1 个赞
这是在连社交平台，配置好发消息机器人不回复，总是报错，问ai说这个强制shell=True windows环境强制shell=False，两个冲突，我解决不了
我也是报错显示这个
这个是 iflow 团队出品的？