[fzf404]

Fixes #10240.

Telegram currently leaves /command@OtherBot unchanged when the target username does not match the current bot. The generic waking stage can then treat the leading / as a wake prefix before command matching, so a command intended for another Telegram bot may wake AstrBot and reach LLM handling.

Modifications / 改动点

• Ignore Telegram commands explicitly addressed to another bot before they enter AstrBot's event pipeline.

• Strip the @username suffix from commands addressed to the current bot.

• Compare Telegram bot usernames case-insensitively.

• Preserve bare commands such as /help, leaving their behavior to the existing AstrBot configuration.

• Add regression tests for current-bot targets, case-insensitive targets, arguments, bare commands, and other-bot targets.

• 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

$ uv run pytest tests/test_telegram_adapter.py -q
37 passed, 1 warning in 4.04s

$ uv run ruff format --check astrbot/core/platform/sources/telegram/tg_adapter.py tests/test_telegram_adapter.py
2 files already formatted

$ uv run ruff check astrbot/core/platform/sources/telegram/tg_adapter.py tests/test_telegram_adapter.py
All checks passed!

The warning is the existing Python 3.12 audioop deprecation warning from tencent_record_helper.py.

---

Checklist / 检查清单

• 😊 The bug and expected routing behavior are documented in [Bug] Telegram 群聊中发给其他机器人的命令会误唤醒 AstrBot #10240.
• 👀 The change is covered by Telegram adapter regression tests and the verification results are included above.
• 📚 No WebUI navigation, page structure, labels, or user-facing documentation changed.
• 🤓 No new dependencies were introduced.
• 😮 This change does not introduce malicious code.

Summary by Sourcery

Prevent Telegram commands intended for other bots from waking AstrBot while preserving existing command behavior.

Bug Fixes:

• Prevent Telegram group commands addressed to other bots from entering AstrBot's event pipeline and potentially triggering LLM handling.
• Normalize commands addressed to the current Telegram bot, including case-insensitive usernames, while preserving bare commands and ordinary mentions.

Tests:

• Add regression coverage for targeted commands, case-insensitive usernames, arguments, bare commands, non-target mentions, private messages, and reply side effects.

[sourcery-ai]

Hey - I've found 1 issue

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="astrbot/core/platform/sources/telegram/tg_adapter.py" line_range="617" />
<code_context>
-                        )
+                    command, bot_name = command_parts[0].split("@", 1)
+                    if bot_name.casefold() != context.bot.username.casefold():
+                        return None
+                    plain_text = command + (
+                        f" {command_parts[1]}" if len(command_parts) > 1 else ""
</code_context>
<issue_to_address>
**issue (broader_impact):** A command addressed to another bot is filtered only after the message's reply chain has already been recursively converted. If the ignored message replies to `/start`, the recursive conversion invokes `self.start()` and sends AstrBot's start message even though the outer command should be ignored.

**Triggers:** When a user sends `/command@OtherBot` as a reply to a Telegram `/start` message.

**Suggested fix:** Normalize and reject explicitly addressed commands before recursively converting the reply message, or suppress side effects while converting reply chains.
</issue_to_address>

Sourcery assessment

Approval pending. 1 finding to address first.

Blocking findings: astrbot/core/platform/sources/telegram/tg_adapter.py:617

---

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

• X
• Mastodon
• LinkedIn
• Facebook

[sourcery-ai]

Sourcery assessment

Approved.

[jamalkamaladdin]

command_token splits on the first space, so a command with a newline before any space merges into bot_name.
A mismatched comparison drops a command addressed to this bot instead of processing it.