openclaw - 💡(How to fix) Fix [Bug]: History compaction strips all user/assistant turns on Qwen, leaving messages[] with only system → upstream 400

Summary

When OpenClaw's history compaction runs on long conversations with Qwen models (Qwen-Long, Qwen-3-Coder-Plus), the outbound messages[] sometimes ends up with no role: user or role: assistant entries — only role: system (or only role: tool) survives.

Qwen upstream rejects with HTTP 400:

• Role must be user or assistant and Content length must be greater than 0
• Role specification invalid

The user-facing symptom (in deployments that route through a proxy that wraps 4xx) is a generic stream-failed message that hides the real error. Without proxy wrapping the operator sees the 400 directly, but the request is always invalid.

Environment

• OpenClaw: 2026.4.2 (planning to upgrade to 2026.5.22)
• Models affected: Qwen-Long (10M context), Qwen-3-Coder-Plus (1M context)
• Provider type: openai-completions compatible
• Trigger threshold: ~20K tokens of history before LCM compaction fires

Repro pattern

1. Start a session with Qwen-Long or Qwen-3-Coder-Plus
2. Run 15-20+ turns with mixed tool use + direct chat to push past the softThreshold
3. Wait for LCM compaction
4. The next user turn → outbound request has messages: [ { role: 'system', ... } ] (and occasionally a trailing role: tool block), with every prior user/assistant turn gone
5. Qwen 400 → request fails

Why this looks distinct from already-tracked compaction issues

Suggested fix direction

After compaction transforms messages[], the serializer should validate that at least one entry has role: 'user' or role: 'assistant' before transmitting. If not, either:

• (a) Re-run compaction with a less-aggressive ratio, preserving the most recent user/assistant pair verbatim, or
• (b) Fall back to a user-visible "history exhausted, please start a new session" error rather than letting upstream return an opaque 400

Option (a) is closer to graceful UX, option (b) is a clearer signal to the operator.

Offered

Happy to provide sanitized session jsonls + outbound request bodies if useful for a regression test. We hit this enough times in production that we can produce captures on demand.