Competitor Intel

OpenClaw: Message Tool Corrupts Transcripts with Delivery-Mirror Side-Branch, Killing Sessions Permanently

Previously, session corruption came from compaction, abort, channel switching, or restart recovery. Now even sending a message mid-turn can permanently kill the session. This expands the failure surface to include every communication action.

📋 Issue Summary

Previously, session corruption came from compaction, abort, channel switching, or restart recovery. Now even sending a message mid-turn can permanently kill the session. This expands the failure surface to include every communication action.

The Message Tool as a Session-Killing Vector

Previously, session corruption came from compaction, abort, channel switching, or restart recovery. Now even sending a message mid-turn can permanently kill the session. This expands the failure surface to include every communication action.

Silent Death, No Recovery

The session dies silently — the agent appears to be working but all subsequent turns fail with HTTP 400. Users have no indication of what went wrong or how to recover, leaving them debugging invisible state corruption.

Eight Independent Mechanisms and Counting

The full catalog now spans abort context loss, incomplete turn poisoning, stale channel fields, abort amnesia, vapor session conflicts, rollover resets, and the delivery-mirror side-branch. Each mechanism is independent, meaning fixing one does not fix the others.

⚠️ Critical Assessment

When the message tool itself can kill sessions, the reliability contract is broken at the most fundamental level. Users cannot trust that sending a message will not permanently destroy their session state. This is not a corner case — it is a core tool creating unrecoverable corruption.

🔗 Sources

Methodology & Sources

This analysis is based on publicly available documentation, community forums (Reddit, Discord, GitHub), vendor-published case studies, security compliance reports, and hands-on testing by the gobii.reviews editorial team. All claims are sourced and verified. We do not accept payment for inclusion or ranking. See our full methodology and editorial standards.

Last updated: June 30, 2026. Published by the gobii.reviews Editorial Team.