Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: OpenCLaw incorrectly parses natural language messages as structured JSON [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#52119•Fetched 2026-04-08 01:15:21
View on GitHub
Comments
1
Participants
2
Timeline
4
Reactions
0
Author
paulhugelpaulhugel
Participants
astroclawastroclawpaulhugelpaulhugel
Timeline (top)
labeled ×2commented ×1subscribed ×1

When a user sends a natural language message containing a word like "Hal" (e.g., "Hal I am in Maui and the timezone is HST"), the OpenCLaw gateway attempts to parse it as structured JSON, resulting in a 400 error:

Error Message

When a user sends a natural language message containing a word like "Hal" (e.g., "Hal I am in Maui and the timezone is HST"), the OpenCLaw gateway attempts to parse it as structured JSON, resulting in a 400 error: 3. Observe a 400 error: The data couldn’t be read because it isn’t in the correct format This results in a 400 error: 2026-03-22T02:33:35.124+00:00 [embedded] run agent end: runId=a2343d93-93f6-4c15-be56-7100272d4b9c isError=true error=400 The data couldn’t be read because it isn’t in the correct format. A 400 status accompanied by “The data couldn’t be read because it isn’t in the correct format” signals that the Jan/OpenClaw container cannot decode the JSON payload it receives or produces when interacting with the Telegram Bot API. Telegram webhooks deliver JSON updates with a defined schema, and Telegram expects well-formed JSON or form-encoded bodies when the bot replies. Verifying the raw request/response bodies, aligning the integration’s data model with Telegram’s current Update object, and checking the container logs and webhook error diagnostics will isolate the malformed payload and resolve the failure.

  • The message text “The data couldn’t be read because it isn’t in the correct format” is the standard exception thrown by Apple’s JSONDecoder when data cannot be decoded into the expected structure, indicating that the OpenClaw/Jan component is failing while decoding Telegram’s JSON update or Telegram’s response.[4]
  • Telegram exposes getWebhookInfo, which reports the most recent delivery error, enabling confirmation of whether Telegram or the local webhook endpoint is producing the 400 response.[2]

Root Cause

Steps to reproduce

  1. Open the OpenCLaw dashboard
  2. Send a message like: "Hal I am in Maui and the timezone is HST"
  3. Observe a 400 error: The data couldn’t be read because it isn’t in the correct format

Fix Action

Fix / Workaround

Actionable Recommendations

  1. Enable verbose logging in the OpenClaw/Jan Telegram connector and review docker logs <container-name> to capture the exact JSON payload around the failure and locate the field that breaks decoding.[5]
  2. Call getWebhookInfo on the bot to read last_error_message and confirm whether Telegram or the local endpoint rejected the payload, narrowing the debugging surface.[2]
  3. Compare the captured JSON against Telegram’s current Update schema and ensure the integration’s decoder treats new or optional fields as nullable/ignored so unexpected data will not abort parsing.[1][3]
  4. When sending replies through Telegram, confirm the request body matches Telegram’s required format (Content-Type header and valid JSON or form encoding) before dispatching, preventing downstream 400 responses.[1]

Code Example

### Redacted OpenCLaw Log Snippet

2026-03-22T02:33:32.443+00:00 [gateway] auth token was missing. Generated a new token and saved it to config (gateway.auth.token).
2026-03-22T02:33:32.772+00:00 [canvas] host mounted at http://127.0.0.1:18789/__openclaw__/canvas/ (root /home/node/.openclaw/canvas)
2026-03-22T02:33:32.798+00:00 [heartbeat] started
2026-03-22T02:33:32.799+00:00 [health-monitor] started (interval: 300s, startup-grace: 60s, channel-connect-grace: 120s)
2026-03-22T02:33:32.800+00:00 [gateway] agent model: anthropic/claude-opus-4-6
2026-03-22T02:33:32.801+00:00 [gateway] listening on ws://127.0.0.1:18789, ws://[::1]:18789 (PID 14)
2026-03-22T02:33:32.802+00:00 [gateway] log file: /tmp/openclaw/openclaw-2026-03-22.log
2026-03-22T02:33:32.810+00:00 [browser/server] Browser control listening on http://127.0.0.1:18791/ (auth=token)
2026-03-22T02:33:33.189+00:00 [gateway] update available (latest): v2026.3.13 (current v2026.3.8). Run: openclaw update
2026-03-22T02:33:34.636+00:00 [bonjour] gateway name conflict resolved; newName="d5edbb3056d8 (OpenClaw) (2)"
2026-03-22T02:33:34.638+00:00 [bonjour] gateway hostname conflict resolved; newHostname="openclaw-(2)"
2026-03-22T02:33:35.123+00:00 [embedded] run agent start: runId=a2343d93-93f6-4c15-be56-7100272d4b9c
2026-03-22T02:33:35.124+00:00 [embedded] run agent end: runId=a2343d93-93f6-4c15-be56-7100272d4b9c isError=true error=400 The data couldn’t be read because it isn’t in the correct format.
RAW_BUFFERClick to expand / collapse

Bug type

Behavior bug (incorrect output/state without crash)

Summary

Bug Report: OpenCLaw incorrectly parses natural language messages as structured JSON

Description

When a user sends a natural language message containing a word like "Hal" (e.g., "Hal I am in Maui and the timezone is HST"), the OpenCLaw gateway attempts to parse it as structured JSON, resulting in a 400 error:

Steps to reproduce

Steps to reproduce

  1. Open the OpenCLaw dashboard
  2. Send a message like: "Hal I am in Maui and the timezone is HST"
  3. Observe a 400 error: The data couldn’t be read because it isn’t in the correct format

Expected behavior

Expected behavior

Natural language messages should be accepted as plain text and processed without attempting to parse them as structured JSON.

For example, when a user says:

  • "I'm in Maui and the timezone is HST"
  • "I'm meeting someone tomorrow"
  • "Can you summarize the meeting?"

...the system should:

  • Accept the message as natural language
  • Process it without attempting to interpret it as a JSON object
  • Respond appropriately in context

Actual behavior

Actual behavior

When a user sends a natural language message containing a word like "Hal" (e.g., "Hal I am in Maui and the timezone is HST"), the OpenCLaw gateway attempts to parse the message as structured JSON (e.g., {"Hal": "I am in Maui and the timezone is HST"}).

This results in a 400 error:
The data couldn’t be read because it isn’t in the correct format

The system fails to process the message due to invalid JSON structure — even though the input is valid natural language.

OpenClaw version

<version 2026.3.8>

Operating system

macOS 26.3.1

Install method

jan application

Model

jan/Jan-v3-4B-base-instruct-4bit

Provider / routing chain

Jan → OpenClaw → Agent → Model Router → AI Provider → AI Model

Additional provider/model setup details

Describe the install method (Jan handled the integration) I did not manually install or configure OpenCLaw. The integration was set up entirely through the Jan application. Steps: 1. Ensured Docker was running 2. Launched the Jan application (version 0.7.8) 3. Went to Settings > Integrations 4. Selected OpenClaw 5. Clicked "Start the OpenClaw Setup" 6. The setup automatically downloaded and launched the OpenCLaw container 7. Selected the Docker image: ghcr.io/openclaw/openclaw:2026.3.8 8. Set the environment variable BRAVE_SEARCH_API_KEY via the Docker environment (passed securely) 9. The OpenCLaw gateway started in a separate Docker container 10. Jan automatically connected to the gateway at http://localhost:8080 (via WebSocket) 11. After setup, the integration was confirmed as active in the Jan interface

Logs, screenshots, and evidence

### Redacted OpenCLaw Log Snippet

2026-03-22T02:33:32.443+00:00 [gateway] auth token was missing. Generated a new token and saved it to config (gateway.auth.token).
2026-03-22T02:33:32.772+00:00 [canvas] host mounted at http://127.0.0.1:18789/__openclaw__/canvas/ (root /home/node/.openclaw/canvas)
2026-03-22T02:33:32.798+00:00 [heartbeat] started
2026-03-22T02:33:32.799+00:00 [health-monitor] started (interval: 300s, startup-grace: 60s, channel-connect-grace: 120s)
2026-03-22T02:33:32.800+00:00 [gateway] agent model: anthropic/claude-opus-4-6
2026-03-22T02:33:32.801+00:00 [gateway] listening on ws://127.0.0.1:18789, ws://[::1]:18789 (PID 14)
2026-03-22T02:33:32.802+00:00 [gateway] log file: /tmp/openclaw/openclaw-2026-03-22.log
2026-03-22T02:33:32.810+00:00 [browser/server] Browser control listening on http://127.0.0.1:18791/ (auth=token)
2026-03-22T02:33:33.189+00:00 [gateway] update available (latest): v2026.3.13 (current v2026.3.8). Run: openclaw update
2026-03-22T02:33:34.636+00:00 [bonjour] gateway name conflict resolved; newName="d5edbb3056d8 (OpenClaw) (2)"
2026-03-22T02:33:34.638+00:00 [bonjour] gateway hostname conflict resolved; newHostname="openclaw-(2)"
2026-03-22T02:33:35.123+00:00 [embedded] run agent start: runId=a2343d93-93f6-4c15-be56-7100272d4b9c
2026-03-22T02:33:35.124+00:00 [embedded] run agent end: runId=a2343d93-93f6-4c15-be56-7100272d4b9c isError=true error=400 The data couldn’t be read because it isn’t in the correct format.

Impact and severity

authored by – Paul Garrett Hugel
ORCID iD: – https://orcid.org/0000-0001-8082-7208

Executive Summary
A 400 status accompanied by “The data couldn’t be read because it isn’t in the correct format” signals that the Jan/OpenClaw container cannot decode the JSON payload it receives or produces when interacting with the Telegram Bot API. Telegram webhooks deliver JSON updates with a defined schema, and Telegram expects well-formed JSON or form-encoded bodies when the bot replies. Verifying the raw request/response bodies, aligning the integration’s data model with Telegram’s current Update object, and checking the container logs and webhook error diagnostics will isolate the malformed payload and resolve the failure.

Evidence-Based Analysis

  • Telegram webhooks always POST a JSON Update object, and outgoing bot calls must supply parameters in JSON or application/x-www-form-urlencoded form; malformed or unexpected fields return HTTP 400 errors from Telegram or any intermediary that cannot parse the payload.[1][3]
  • The message text “The data couldn’t be read because it isn’t in the correct format” is the standard exception thrown by Apple’s JSONDecoder when data cannot be decoded into the expected structure, indicating that the OpenClaw/Jan component is failing while decoding Telegram’s JSON update or Telegram’s response.[4]
  • Telegram exposes getWebhookInfo, which reports the most recent delivery error, enabling confirmation of whether Telegram or the local webhook endpoint is producing the 400 response.[2]
  • Inspecting the Docker container logs (e.g., docker logs <container-name>) captures the raw payloads and stack trace, helping pinpoint which field or encoding causes the decode failure.[5]

Actionable Recommendations

  1. Enable verbose logging in the OpenClaw/Jan Telegram connector and review docker logs <container-name> to capture the exact JSON payload around the failure and locate the field that breaks decoding.[5]
  2. Call getWebhookInfo on the bot to read last_error_message and confirm whether Telegram or the local endpoint rejected the payload, narrowing the debugging surface.[2]
  3. Compare the captured JSON against Telegram’s current Update schema and ensure the integration’s decoder treats new or optional fields as nullable/ignored so unexpected data will not abort parsing.[1][3]
  4. When sending replies through Telegram, confirm the request body matches Telegram’s required format (Content-Type header and valid JSON or form encoding) before dispatching, preventing downstream 400 responses.[1]

Sources
[1] https://core.telegram.org/bots/api#making-requests
[2] https://core.telegram.org/bots/api#getwebhookinfo
[3] https://core.telegram.org/bots/api#update
[4] https://developer.apple.com/documentation/foundation/jsondecoder
[5] https://docs.docker.com/engine/reference/commandline/logs/

Additional information

No response

extent analysis

Fix Plan

To resolve the issue of OpenCLaw incorrectly parsing natural language messages as structured JSON, we need to modify the JSON decoding logic to handle messages that are not in JSON format. Here are the steps:

  • Update the OpenClaw gateway to detect if the incoming message is in JSON format before attempting to parse it.
  • If the message is not in JSON format, process it as a natural language message.
  • Modify the Telegram connector to handle non-JSON messages and prevent 400 errors.

Code Changes

Here is an example of how you can modify the JSON decoding logic in OpenClaw:

import json

def process_message(message):
    try:
        # Attempt to parse the message as JSON
        json_message = json.loads(message)
        # Process the JSON message
        process_json_message(json_message)
    except json.JSONDecodeError:
        # If the message is not in JSON format, process it as a natural language message
        process_natural_language_message(message)

def process_natural_language_message(message):
    # Implement logic to process natural language messages
    pass

def process_json_message(json_message):
    # Implement logic to process JSON messages
    pass

Temporary Workaround

If you are unable to modify the OpenClaw gateway, you can use a temporary workaround to prevent 400 errors. You can add a check in the Telegram connector to detect if the message is in JSON format before sending it to the OpenClaw gateway. If the message is not in JSON format, you can modify it to be in JSON format or handle it differently.

Verification

To verify that the fix worked, you can test the OpenClaw gateway with different types of messages, including natural language messages and JSON messages. You can also check the logs to ensure that the 400 errors are no longer occurring.

Extra Tips

  • Make sure to handle errors and exceptions properly to prevent crashes and unexpected behavior.
  • Consider adding logging and monitoring to detect and diagnose issues with the OpenClaw gateway and Telegram connector.
  • Keep the OpenClaw gateway and Telegram connector up to date with the latest versions and security patches.

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

Natural language messages should be accepted as plain text and processed without attempting to parse them as structured JSON.

For example, when a user says:

  • "I'm in Maui and the timezone is HST"
  • "I'm meeting someone tomorrow"
  • "Can you summarize the meeting?"

...the system should:

  • Accept the message as natural language
  • Process it without attempting to interpret it as a JSON object
  • Respond appropriately in context
#api#environment variable#model compatibility#GPU setup#container setup

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]: OpenCLaw incorrectly parses natural language messages as structured JSON [1 comments, 2 participants]