Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug] Self-messages invoke agent model despite selfChatMode: false [1 comments, 2 participants]

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…
GitHub stats
openclaw/openclaw#69854•Fetched 2026-04-22 07:47:25
View on GitHub
Comments
1
Participants
2
Timeline
3
Reactions
0
Author
openclawcria-criaopenclawcria-cria
Participants
openclawcria-criaopenclawcria-criarafiki270rafiki270
Timeline (top)
labeled ×2commented ×1

With selfChatMode: false, outbound WhatsApp messages sent by the agent are still passed to the agent model for evaluation, causing unnecessary API token consumption (~$1+/day idle cost).

Root Cause

With selfChatMode: false, outbound WhatsApp messages sent by the agent are still passed to the agent model for evaluation, causing unnecessary API token consumption (~$1+/day idle cost).

RAW_BUFFERClick to expand / collapse

Bug type

Regression (worked before, now fails)

Beta release blocker

No

Summary

With selfChatMode: false, outbound WhatsApp messages sent by the agent are still passed to the agent model for evaluation, causing unnecessary API token consumption (~$1+/day idle cost).

Steps to reproduce

  1. Set selfChatMode: false in openclaw.json channels.whatsapp
  2. Trigger any workflow that sends a WhatsApp reply
  3. Observe in gateway logs: the outbound message is received back as inbound (self) and passed to the agent model

Expected behavior

With selfChatMode: false, self-messages should be filtered at the gateway level before reaching the agent model. No API call should be made.

Actual behavior

Agent model (Haiku) is invoked to evaluate each self-message. Gateway logs show web-auto-reply processing every outbound message as inbound. Costs ~$1+/day in unnecessary token usage.

OpenClaw version

2026.4.15

Operating system

Ubuntu 24

Install method

Docker

Model

anthropic/claude-haiku-4-5

Provider / routing chain

openclaw -> anthropic (direct API key)

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

Affected: all WhatsApp installations with active workflows Severity: Medium (idle cost drain, not a functional failure) Frequency: Every workflow execution Consequence: ~$1+/day in unnecessary API costs on a moderately active installation

Additional information

No response

extent analysis

TL;DR

Update the gateway configuration to filter self-messages when selfChatMode is set to false to prevent unnecessary API token consumption.

Guidance

  • Review the openclaw.json configuration file to ensure that the selfChatMode setting is correctly applied to the WhatsApp channel.
  • Verify that the gateway is properly configured to filter self-messages based on the selfChatMode setting.
  • Check the workflow triggers to ensure that they are not inadvertently sending outbound messages back to the agent model for evaluation.
  • Investigate the possibility of updating the openclaw version to a newer release, as the current version (2026.4.15) may contain a bug or outdated configuration.

Example

No code snippet is provided as the issue is related to configuration and workflow setup.

Notes

The issue is specific to WhatsApp installations with active workflows and selfChatMode set to false. The severity is medium, and the consequence is a daily cost drain of ~$1+ due to unnecessary API token consumption.

Recommendation

Apply a workaround by updating the gateway configuration to filter self-messages when selfChatMode is set to false, as this is a configuration-related issue rather than a version-specific problem.

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

With selfChatMode: false, self-messages should be filtered at the gateway level before reaching the agent model. No API call should be made.

#api#installation#mixed precision#training loop#device allocation

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 [Bug] Self-messages invoke agent model despite selfChatMode: false [1 comments, 2 participants]