openclaw - 💡(How to fix) Fix Telegram progress updates truncate useful work-log context

Problem

Telegram tool/progress updates currently truncate explanatory progress messages aggressively. In a real direct-chat workflow, updates such as:

• I found the prior notes: we had done a first-pass...
• Shell access is working in this fresh turn...
• gh is not installed, and the repo is not already cloned...
• Homebrew is auto-updating before installing gh...

show up clipped before the useful part lands. This hides the agent's visible work-log: what it checked, what it found, why it matters, and what happens next.

This is not a request to expose hidden model reasoning. It is about preserving user-authored/assistant-authored visible progress summaries that are already intended for the user.

Desired behavior

Add or tune a Telegram progress mode that keeps tool spam compact while allowing longer explanatory progress lines.

Possible shape:

• Keep raw command/tool details collapsed or summarized.
• Allow progress/status text around 180-240 chars instead of ~96 chars.
• Prefer complete sentence truncation where possible.
• Preserve exact blocker/error messages when short enough.

Why it matters

For long-running agent tasks, especially repo audits and installs, these updates are the user's main confidence signal. Over-truncation makes the agent look opaque even when it is trying to explain its work.

Example context

During a wystack-agent-kit review, the Telegram progress stream showed clipped entries like:

I found the prior notes: we had done a first-pass...concrete advice and fixes/priorities.
Shell access is working in this fresh turn, so the earlier hook-relay block is gone. Next I’m...
'gh' is not installed, and the repo is not already cloned under the usual places. I found the...
Homebrew is auto-updating before installing 'gh', so this may take a minute. After that I’ll...

The requested improvement is a roomier visible work-log, not hidden chain-of-thought exposure.