fix: waking bot with a reply component and empty user prompt

[lingyun14beta]

fix #8368
当用户引用任意消息并 @ Bot 但不附加文字时，消息链为 [Reply, At]，message_str 为空，导致以下两个问题：

1. internal.py 在 message_str 为空时直接跳过 LLM 请求，即使消息链中包含 Reply 组件。
2. astr_main_agent.py 在 _decorate_llm_request 执行之前就检查内容是否为空，此时引用消息文字尚未注入 extra_user_content_parts，导致 build_main_agent 返回 None。

Modifications / 改动点

• astrbot/core/pipeline/process_stage/method/agent_sub_stages/internal.py：新增 has_reply 判断，消息链含 Reply 组件时不跳过 LLM 请求。
• astrbot/core/astr_main_agent.py：将空内容检查移至 _decorate_llm_request 之后，确保引用消息内容已注入后再判断；同时去掉群聊限制条件 not event.get_group_id()。

• This is NOT a breaking change. / 这不是一个破坏性变更。

Screenshots or Test Results / 运行截图或测试结果

正常回复

---

Checklist / 检查清单

• 😊 If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc.
  / 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。

• 👀 My changes have been well-tested, and "Verification Steps" and "Screenshots" have been provided above.
  / 我的更改经过了良好的测试，并已在上方提供了“验证步骤”和“运行截图”。

• 🤓 I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in requirements.txt and pyproject.toml.
  / 我确保没有引入新依赖库，或者引入了新依赖库的同时将其添加到 requirements.txt 和 pyproject.toml 文件相应位置。

• 😮 My changes do not introduce malicious code.
  / 我的更改没有引入恶意代码。

Summary by Sourcery

Ensure LLM requests are triggered when users reply to a message and mention the bot without additional text.

Bug Fixes:

• Prevent LLM requests from being skipped when the message contains only a reply and an @bot mention with no extra text.
• Allow empty prompts to be replaced with a placeholder when there is a reply or extra user content, instead of aborting the main agent build.

[sourcery-ai]

Hey - I've left some high level feedback:

• In internal.py, the new has_reply condition will cause any reply-only message (even when the replied content is empty or non-user generated) to proceed to LLM; consider whether you should additionally check the replied message type/source to avoid unnecessary LLM invocations.
• After moving _decorate_llm_request before the emptiness check in build_main_agent, verify that _decorate_llm_request is safe to call when req is initially empty and that it cannot introduce side effects (e.g., long-running operations) that should only occur when a real LLM request will be sent.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- In `internal.py`, the new `has_reply` condition will cause any reply-only message (even when the replied content is empty or non-user generated) to proceed to LLM; consider whether you should additionally check the replied message type/source to avoid unnecessary LLM invocations.
- After moving `_decorate_llm_request` before the emptiness check in `build_main_agent`, verify that `_decorate_llm_request` is safe to call when `req` is initially empty and that it cannot introduce side effects (e.g., long-running operations) that should only occur when a real LLM request will be sent.

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[gemini-code-assist]

Code Review

This pull request introduces support for handling reply messages in the LLM request pipeline and adjusts the logic for detecting empty messages. However, moving _decorate_llm_request before the empty prompt check introduces a bug where system reminders are appended to extra_user_content_parts, causing empty messages to erroneously trigger LLM requests. It is recommended to keep the original execution order and instead check for Reply components directly during the empty prompt validation.

[lingyun14beta]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces support for handling message replies in the main agent and pipeline stages. A critical bug was identified in astrbot/core/astr_main_agent.py where replacing the check for req.extra_user_content_parts with has_reply would cause the bot to ignore other attachments like files or videos when there is no reply. The reviewer suggested combining both conditions and adding unit tests.

[lingyun14beta]

/gemini review

[gemini-code-assist]

Code Review

This pull request introduces support for handling Reply message components. It ensures that messages containing replies are not skipped as empty messages and sets the prompt to <attachment> when appropriate. There are no review comments, and I have no additional feedback to provide.

[Blueteemo]

@sourcery-ai review

[sourcery-ai]

Hi @Blueteemo! 👋

Only authors and team members can run @sourcery-ai commands on public repos.

[zouyonghe]

@sourcery-ai review

[sourcery-ai]

Hey - I've left some high level feedback:

• The has_reply detection logic is now duplicated in both astr_main_agent.build_main_agent and internal.process; consider extracting a shared helper (e.g., message_has_reply(event.message_obj.message)) so future changes to what counts as a non-empty message (like additional reply-like components) stay consistent across the pipeline.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- The `has_reply` detection logic is now duplicated in both `astr_main_agent.build_main_agent` and `internal.process`; consider extracting a shared helper (e.g., `message_has_reply(event.message_obj.message)`) so future changes to what counts as a non-empty message (like additional reply-like components) stay consistent across the pipeline.

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}