Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: Codex-native web_search is mirrored into session/chat logs with empty query and stub-only result [1 pull requests]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

When using the Codex/OpenAI native web search path, OpenClaw mirrors web_search tool calls into the session/chat transcript with an empty query and a generic completed result.

Root Cause

This is misleading because the model clearly performed a substantive search, but the stored tool card/log implies it searched for an empty string.

Fix Action

Fixed

RAW_BUFFERClick to expand / collapse

Bug type

Behavior bug (incorrect output/state without crash)

Beta release blocker

No

Summary

When using the Codex/OpenAI native web search path, OpenClaw mirrors web_search tool calls into the session/chat transcript with an empty query and a generic completed result.

Steps to reproduce

Repro

Use an OpenClaw session on openai-codex-responses
Ask for current information that triggers native web search
Inspect the resulting tool card or session .jsonl
Observe web_search mirrored with {"query":""} and stub-only result

Expected behavior

  • preserve actual query text when available
  • if unavailable, use an explicit sentinel instead of ""
  • preserve more result metadata than just {"status":"completed"}

Actual behavior

Codex-native web_search is mirrored with empty query in session/chat logs

When using openai-codex-responses, OpenClaw mirrors native web_search calls into the session transcript as {"query":""} with a stub result {"status":"completed"}.

This is misleading because the model clearly performed a substantive search, but the stored tool card/log implies it searched for an empty string.

OpenClaw version

2026.5.19

Operating system

ubuntu 24.04

Install method

npm

Model

openai codex

Provider / routing chain

openclaw -> openai/gpt-5.4-codex

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

severity low, just makes openclaw more opaque as you dont know what it was doing.

Additional information

No response

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

  • preserve actual query text when available
  • if unavailable, use an explicit sentinel instead of ""
  • preserve more result metadata than just {"status":"completed"}

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING