Summary

Describe the problem and fix in 2–5 bullets:

• Problem: Anthropic compatible endpoints would not work after setup.
• Why it matters: Some of us want to run models locally!
• What changed: The onboarding wizard differs from the rest of the application, in that it uses a raw fetch call rather than the anthropic client to contact the LLM server. The onboarding wizard now injects a v1 for making the raw fetch call to verify the endpoint. The rest of the application uses the anthropic client, which does this automatically.
• What did NOT change (scope boundary): nothing else changed. This will only affect onboarding, so existing configs will not be touched.

Change Type (select all)

• [ x] Bug fix
• Feature
• Refactor
• Docs
• Security hardening
• Chore/infra

Scope (select all touched areas)

• Gateway / orchestration
• Skills / tool execution
• Auth / tokens
• Memory / storage
• Integrations
• [ x] API / contracts
• UI / DX
• CI/CD / infra

Linked Issue/PR

• Closes # [Bug]: Custom Provider Anthropic-Compatible 404 Error Due to Mismatch /v1 Behavior in Setup and Usage #21261
• Related #

User-visible / Behavior Changes

List user-visible changes (including defaults/config).
None.

Security Impact (required)

• New permissions/capabilities? (Yes/No) No
• Secrets/tokens handling changed? (Yes/No) No
• New/changed network calls? (Yes/No) Yes, fixes calls to Anthropic compatible APIs.
• Command/tool execution surface changed? (Yes/No) No.
• Data access scope changed? (Yes/No) No.
• If any Yes, explain risk + mitigation:

Tiny change to onboarding wizard only to match behavior of the rest of the application, which uses the anthropic client. The onboarding wizard uses a raw fetch, requiring us to manually inject /v1.

Repro + Verification

If you run locally, should be straightforward to test it works. Just add a custom provider with anthropic compatible selected, and see if works right off the bat.

Environment

• OS: MacOS
• Runtime/container:
• Model/provider: Ollama
• Integration/channel (if any):
• Relevant config (redacted):

Steps to reproduce

1. Configure something like Ollama manually as a custom provider, selecting Anthropic compatible. The onboarding wizard will happily confirm that something like localhost:1234/v1 is a very valid and successful endpoint.
2. Try using OpenClaw tui, you'll get a 404 not found error since it's hitting an invalid endpoint of /v1/v1/messages. Manually removing the /v1 from the config file will fix it.

Expected behavior

The setup script and regular usage of anthropic compatible providers should have identical behavior for calling models.

Actual behavior

Instead it appears they are different.

Evidence

Attach at least one:

• Failing test/log before + passing after
• [x ] Trace/log snippets
• Screenshot/recording
• Perf numbers (if relevant)

Logs from attaching a listener between Ollama and OpenClaw:

..z7Q:..POST /v1/v1/messages HTTP/1.1

Human Verification (required)

What you personally verified (not just CI), and how:

• Verified scenarios:
• Edge cases checked:
• What you did not verify: Still need to run locally and confirm it works.

Compatibility / Migration

• Backward compatible? (Yes/No) Yes
• Config/env changes? (Yes/No) Kind of? Tiny tweak for onboarding wizard of custom providers.
• Migration needed? (Yes/No) No.
• If yes, exact upgrade steps:

Failure Recovery (if this breaks)

• How to disable/revert this change quickly: Contributors can easily rollback, and users can manually go into their config files to adjust custom provider URLs.
• Files/config to restore:
• Known bad symptoms reviewers should watch for:

Risks and Mitigations

List only real risks for this PR. Add/remove entries as needed. If none, write None.

• Risk: None
  • Mitigation:

Greptile Summary

Fixed Anthropic-compatible endpoint configuration in onboarding wizard. The wizard now conditionally appends /v1 to the base URL when making verification requests, matching the behavior of the Anthropic client used in the rest of the application. This prevents double /v1/v1/messages paths that caused 404 errors when using Anthropic-compatible providers like Ollama. The fix only affects the onboarding wizard verification step; existing configurations are not modified.

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• The change is a focused bug fix that adds conditional /v1 path handling only for the onboarding verification step. The fix is well-documented with an inline comment referencing the configuration docs, uses a simple regex pattern to check for existing /v1 suffix, and aligns with the documented behavior that base URLs should omit /v1 since the Anthropic client appends it automatically. The scope is limited to the verification function used during onboarding, leaving existing configurations untouched.
• No files require special attention

_{Last reviewed commit: b7fb46f}

_{(2/5) Greptile learns from your feedback when you react with thumbs up/down!}