Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

[Bug]: Native Gateway Local AI setup fails its response test despite retries

Open
  • Is something wrong?
#1,713 1 comment 1 reaction 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
45/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Active
Tech stack
csharp
Domain
backend

Research direction

Start with the Windows Companion's native Gateway setup and real-response verification flow, then inspect the recovery panel and diagnostic logging. The report does not name files or tests, so first locate where the generic activation error is shown and where Gateway failures are recorded. Use the reporter's Companion and Gateway versions and logs to identify the cause; done means the failure is diagnosable and setup either completes or shows the underlying error and a useful log location.

Written by the indexing model from the issue text.

Description

impact:ux-release-blocker issue-rating: 🦪 silver shellfish P0

Bug

A user reports that Local AI setup through the recommended native Gateway still fails after retrying. The user's screenshot reaches the real-response verification step and reports: "The provider could not activate this model. Refresh to choose again."

Reported reproduction

  1. Use the latest Windows Companion and select the recommended native Gateway route.
  2. Configure Local AI and let setup test a real response from the selected model through the Gateway.
  3. Retry after the failure. The reporter confirms that the failure persists.

Actual behavior

The real-response test needs attention. Earlier progress stages show Complete. The Gateway recovery panel offers Open terminal, Restart Gateway, and Start AI setup again, but the visible error does not identify the underlying failure.

An earlier screenshot from the same reporter also shows native Gateway startup exiting with code 78. Whether these failures share a cause is unknown.

Expected behavior

Local AI setup completes a real model response through the native Gateway. If verification fails, recovery provides the underlying error and a useful diagnostic log location.

Evidence and limits

  • Reported on the latest Windows Companion; exact Companion/Gateway versions, model ID, Windows build, and logs are not yet available.
  • Not reproduced locally: native onboarding on the Windows 11 ARM64 test machine is blocked by its live capability check.
  • An outdated Gateway causing this failure remains a hypothesis.

Reporter screenshot showing the final response-test failure and Gateway recovery actions:

Local AI setup reaches the real-response test, which needs attention, and opens Gateway recovery

Follow-up failure screenshot accompanying confirmation that retrying did not recover Local AI:

Follow-up Local AI failure with the same generic provider activation error and Gateway recovery actions

Related

  • #1583 ([Bug]: Native Gateway + Local AI chat fails with DataCloneError; Companion stays on "Assistant is thinking...") concerns chat after setup verification succeeds. This report fails during setup verification.
  • #1672 ([P0][Bug]: Native Store Gateway onboarding requires manual clawctl setup before the wizard can continue) concerns profile preparation, an earlier stage. Neither is confirmed to be the same root cause.
  • #1242 (Local AI setup intermittently fails (~30%): llama-server CUDA crash (cudaErrorSharedObjectInitFailed in flash-attn) on RTX Spark N1X during warm-up) documents a CUDA/model warm-up failure. This screenshot marks the earlier model-start check Complete. Logs are needed to determine whether a later runtime failure is related.
  • #1578 (during MSIX onboarding, fails if gateway is already present) shows WinGet certificate error 0x8A15005E during package acquisition, before Local AI verification.

No exact duplicate was identified among the open reports reviewed. A shared underlying cause remains possible until the reporter's versions and logs are collected.

Dominant language
C#
Stars
2.2k
Forks
321
Avg merge
3d 17h
Merged PRs (30d)
173

Getting set up

  • No Dockerfile or Docker Compose file
  • Has a pull request template
  • No contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from openclaw/openclaw-windows-node

All issues in openclaw/openclaw-windows-node

Similar issues

More C# issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.