Keep the chosen provider on a model name that contains a slash

[kevin9327]

What this changes

A model name the endpoint publishes with a slash in it — qwen/qwen3-8b, or the docs example openai/gpt-5.6-terra — was treated as already carrying a provider. LiteLLM then took the first half as the provider and sent only the rest, so a compatibility endpoint either failed with LLM Provider NOT provided or received gpt-5.6-terra instead of the namespaced name.

The ADK, Strands, Agno, LlamaIndex and CrewAI Bots now always keep the chosen provider in front. LiteLLM strips that first component when it talks to the endpoint, so the whole model name arrives.

I will rebase after merge if CHANGELOG conflicts.

Where it runs

OpenBot is deployed as several server processes behind a load balancer, serving a whole company.
Consecutive requests from the same person reach different processes, and the process that answered a
WebSocket upgrade is rarely the one that answers the next call on that conversation.

State that outlives a single request therefore has to be shared, or the change works on one machine
and stops working the moment there are two, without saying so. That failure is worse than not
shipping the feature: it passes review, passes CI, passes a local demo, and only surfaces as a Bot
that forgets, a question nobody can answer, or a boundary that never fires.

Answer these even when the answer is "none":

• New state that outlives a request? None. Each harness still reads BOT_PROVIDER and BOT_MODEL from the environment when it starts.
• What happens on the second replica? The same. Every replica of a harness process is started with the same two variables.
• Anything serialised? None.
• Anything fanned out to a browser? None.
• New listener, port, or schedule? None.

Boundary and audit

• Every acting call still goes through the gateway: resolve, decide, audit, then act. Untouched.
• New refusals and new failures each write a row. None added.
• Nothing new is trusted from the client that the server can resolve itself. The model name is still the one the setup screen wrote.

Changelog

• A line in CHANGELOG.md under Unreleased.

Proof

On unmodified main, with the new tests present:

• CrewAI _model(): qwen/qwen3-8b stayed qwen/qwen3-8b; openai/gpt-5.6-terra stayed openai/gpt-5.6-terra.
• Strands and ADK roundtrips: qwen/qwen3-8b failed with LLM Provider NOT provided; openai/gpt-5.6-terra was sent to the endpoint as gpt-5.6-terra.

After the fix, those same tests pass. The fake OpenAI server records qwen/qwen3-8b and openai/gpt-5.6-terra as the model names. Existing model-choice cases still pass (Agno 5, LlamaIndex 5, ADK 5, Strands 7, CrewAI unit 10). bun run typecheck clean. Biome format:check and lint clean.

AI-assisted (Grok)

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.