zeroclaw-labs/zeroclaw · warning · DiagItem

{label}: no model configured

Error message

{label}: no model configured

What it means

check_config_semantics warns when a provider profile has no `model` configured. Without a model name on the profile, requests routed through it cannot resolve a concrete model and will fail or behave unexpectedly downstream.

Source

Thrown at crates/zeroclaw-runtime/src/doctor/mod.rs:1164

            }

            // API key presence
            if family != "ollama" {
                if entry.api_key.as_deref().is_some() {
                    items.push(DiagItem::ok(cat, format!("{label}: API key configured")));
                } else {
                    items.push(DiagItem::warn(
                        cat,
                        format!("{label}: no api_key set (may rely on env vars or model_provider defaults)"),
                    ));
                }
            }

            // Model configured
            if let Some(model) = entry.model.as_deref() {
                items.push(DiagItem::ok(cat, format!("{label}: model: {model}")));
            } else {
                items.push(DiagItem::warn(cat, format!("{label}: no model configured")));
            }

            // A missing value remains unknown until this profile is selected;
            // zero is explicitly invalid because it leaves recovery with no
            // usable model-context budget.
            match entry.context_window {
                Some(0) => items.push(DiagItem::error(
                    cat,
                    crate::i18n::get_required_cli_string_with_args(
                        "cli-doctor-context-window-zero",
                        &[("provider_ref", &label)],
                    ),
                )),
                Some(context_window) => items.push(DiagItem::ok(
                    cat,
                    crate::i18n::get_required_cli_string_with_args(
                        "cli-doctor-context-window-ok",
                        &[

View on GitHub (pinned to 88bb9c8533)

Solutions

1. Set `model` on the provider profile to a concrete model name (e.g. `gpt-4o-mini` or the local model id)
2. If model routes are meant to override the model, still set a safe default on the profile
3. Re-run `zeroclaw doctor` to confirm the profile reports `model: ...` as ok

Example fix

# before
[providers.openai]
family = "openai"
api_key = "sk-..."

# after
[providers.openai]
family = "openai"
api_key = "sk-..."
model = "gpt-4o-mini"

import tomllib

cfg = tomllib.load(open('zeroclaw.toml', 'rb'))
for name, p in cfg.get('providers', {}).items():
    assert p.get('model'), f'{name}: no model configured'

When it happens

Trigger: A providers entry with an api_key but no `model` field; doctor prints `{label}: no model configured` during any diagnostic run.

Common situations: Profiles created with only auth in mind, expecting routes to supply the model; config migration dropping a renamed field; multi-model setups missing the default.

Related errors

• model route "{}" has empty model
• model route "{}" uses invalid model_provider "{}": {}
• embedding route "{}" uses invalid model_provider "{}": {}
• agent "{name}" uses invalid model_provider "{provider_ref}":
• config section `{$path}` is malformed and was reset to defau

AI-assisted analysis of zeroclaw-labs/zeroclaw@88bb9c8533 (2026-08-23). Data as JSON: /api/errors/f884357d2901bd6f. Report an issue: GitHub.