/history 命令硬编码截断消息到 200 字符，导致回复显示不完整。/history command hardcodes the truncated message to 200 characters, resulting in an incomplete response display

[Saillllllll]

Feature Area

Core / Engine

Problem or Motivation

/history 命令回放的消息内容被硬编码截断到 200 个字符（rune），导致长回复显示不全。Agent 回复通常远超 200 字符，/history 只能看到前一小段
正常实时回复没有此限制，导致 /history 的回放质量远不如实时消息。在移动端接续桌面端工作时回溯历史消息非常不利，比如在桌面端发出了一个命令，想在移动端查看结果时，无法看到完整结果。The message content replayed by the /history command is hard-coded to be truncated to 200 characters (runes), which causes long replies to be displayed incompletely. Agent replies are often much longer than 200 characters, so /history only shows the first short segment. Normal real-time replies do not have this limitation, which makes /history playback quality much worse than live messages.

Proposed Solution

将 200 改为可配置项（如 [display] 中 history_max_len，0=不截断）或提高默认值到 2000，并设为可配置。Make 200 a configurable option, for example history_max_len under [display], where 0 means no truncation. Alternatively, increase the default value to 2000 and make it configurable.

Alternatives Considered

无。None。

Additional Context

源码位置

core/engine.go 中两处都存在此问题：

1. cmdHistory（文本模式，L7217-7220）

content := h.Content
if len([]rune(content)) > 200 {
    content = string([]rune(content)[:200]) + "..."
}

2. renderHistoryCard（卡片模式，L10758-10760）

content := h.Content
if len([]rune(content)) > 200 {
    content = string([]rune(content)[:200]) + "..."
}

Source Code Location

This issue exists in two places in core/engine.go:

1. cmdHistory（text mode，L7217-7220）

content := h.Content
if len([]rune(content)) > 200 {
    content = string([]rune(content)[:200]) + "..."
}

2. renderHistoryCard（card mode，L10758-10760）

content := h.Content
if len([]rune(content)) > 200 {
    content = string([]rune(content)[:200]) + "..."
}

Willingness to Contribute

• I'd be willing to submit a PR for this feature

[github-actions]

👋 Hi @Saillllllll! Thanks for opening your first issue here!

We'll review it as soon as possible. While waiting, please make sure:

• You've provided enough details about the issue
• Any error messages are included in full
• For feature requests, describe your use case

Thanks for your feedback!

[chenhg5]

Good catch! The 200-character truncation is indeed hardcoded. A more reasonable approach would be to make this configurable or at least increase the default significantly (e.g. 500–1000 characters) so long AI replies aren't cut off in the history view.

We'll address this in an upcoming release. For now, the workaround is to use /current which shows the full current session, or use the Web Admin UI to browse full conversation history.

[AaronZ345]

Opened PR #1291 to fix this: #1291

The patch adds [display].history_max_len with project-level override support, raises the default per-entry /history limit from the hard-coded 200 chars to 1000, and allows history_max_len = 0 to disable truncation. Both text /history output and history cards use the same helper now.

Validation: GitHub CI is all green (lint, unit-test, smoke-test, regression-test, performance-test).