openclaw - 💡(How to fix) Fix read tool output truncated for medium-sized files (2026.5.18 regression)

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…

The read tool truncates file content at a lower threshold than expected after the 2026.5.18 update. When reading a 365-line / 8737-byte SKILL.md file, the output appeared incomplete in the agent's context — content that was fully returned by the tool was either truncated by the context window or by a new tool output cap introduced in the update.

Error Message

  • No explicit error message — content was simply not delivered in full

Root Cause

The read tool truncates file content at a lower threshold than expected after the 2026.5.18 update. When reading a 365-line / 8737-byte SKILL.md file, the output appeared incomplete in the agent's context — content that was fully returned by the tool was either truncated by the context window or by a new tool output cap introduced in the update.

RAW_BUFFERClick to expand / collapse

Summary

The read tool truncates file content at a lower threshold than expected after the 2026.5.18 update. When reading a 365-line / 8737-byte SKILL.md file, the output appeared incomplete in the agent's context — content that was fully returned by the tool was either truncated by the context window or by a new tool output cap introduced in the update.

Impact

  • Agents cannot reliably read medium-sized files (even <10KB) and get the full content
  • Task-relevant information may be silently lost
  • Debugging and configuration-reading workflows are degraded

Additional context

This may be related to the 2026.5.18 change: "Agents/tools: shorten built-in tool descriptions and schema hints" — if the read tool's schema hint was shortened in a way that affects how the agent or runtime handles output, that could explain the regression. Alternatively, the update may have introduced a hard output cap on tool results.

Environment

  • OpenClaw version: 2026.5.18 (upgraded from 2026.5.17)
  • File read: 365 lines, 8737 bytes (well within typical limits)
  • No explicit error message — content was simply not delivered in full

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…

Still need to ship something?

×6

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

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix read tool output truncated for medium-sized files (2026.5.18 regression)