Skip to content

[Bug]: WebChat keeps a visible 'Tool output' block after the assistant reply is complete #89781

Description

@lamyliu

Summary

In WebChat direct chat, a Tool output UI block can remain visible in the transcript even after the assistant has already produced a normal final answer.

This looks like an internal tool-state/rendering artifact leaking into the user-facing chat UI. In the screenshot, the assistant answer is already complete, but a separate Tool output card still appears below it.

Impact

  • Confusing user experience
  • Makes the conversation look unfinished or broken
  • Exposes implementation details (Tool output) that should likely stay hidden in normal chat UX
  • Can stack with the earlier commentary-leak issue and further reduce trust

Environment

  • OpenClaw: 2026.5.28 (e932160)
  • Surface/channel: webchat direct chat
  • Runtime model: openai-codex/gpt-5.4
  • Observed time: Wed 2026-06-03 17:42 GMT+8

What happened

After the assistant answered a user’s question normally, the WebChat UI still showed a separate expandable row labeled Tool output under the assistant reply.

The user asked why the tool output was still there because it looked like the tool/render state had not been cleaned up.

Expected behavior

After a normal final answer is delivered, internal tool output containers should be hidden, collapsed into developer-only/debug-only state, or otherwise not shown in the end-user transcript.

Actual behavior

A visible Tool output row remains in the user-facing chat transcript after the reply.

Repro notes

Observed in a WebChat direct conversation after a turn that used tools and then returned a normal final response.

Rough flow:

  1. User asks for interpretation of a news link
  2. Assistant uses tools
  3. Assistant returns a normal final answer
  4. WebChat still shows a Tool output block below the answer

Screenshot

Persistent Tool output block in WebChat

Direct link if embed does not render:

Notes

This may be related to the broader class of WebChat rendering/state cleanup issues where internal tool activity is leaking into user-visible UI. But this specific issue is about the persistent Tool output block remaining visible after the answer is complete.

Metadata

Metadata

Assignees

No one assigned

    Labels

    P3Low-priority cleanup, docs, polish, ergonomics, or speculative work.clawsweeper:fix-shape-clearClawSweeper found a clear likely implementation shape for this issue.clawsweeper:needs-maintainer-reviewClawSweeper marked this issue as needing maintainer review before automation.clawsweeper:needs-product-decisionClawSweeper marked this issue as needing a product or behavior decision.clawsweeper:no-new-fix-prClawSweeper does not recommend queueing a new automated fix PR for this issue.clawsweeper:source-reproClawSweeper found a high-confidence source-level issue reproduction.impact:otherThis issue has meaningful maintainer-visible impact outside the owned taxonomy.issue-rating: 🦞 diamond lobsterVery strong issue quality with high-confidence source-level or clear reproduction.

    Type

    No type

    Fields

    Priority

    None yet

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions