Avoid syncing model-routed providers as current

When a model route matches, this replaces the provider list with only the route target, but the later success path still builds SuccessSyncInfo and calls ProxyServerState::sync_successful_provider_selection, which persists set_current_provider and may update the live backup whenever the route target differs from current_provider_id_at_start. In a setup where *sonnet* routes to provider B while the user’s current provider is A, the first successful routed request permanently switches the app to B, so subsequent non-routed traffic and live config no longer use the user-selected provider.

Useful? React with 👍 / 👎.

Anchor wildcard model route patterns

Wildcard patterns are compiled without ^...$, so regex matching can succeed on any substring rather than the whole model id. For example, a user adding the documented prefix route claude-* would also match xclaude-opus, and sonnet* would match claude-sonnet even though no leading wildcard was supplied; this makes route precedence unpredictable when model ids merely contain another route’s prefix.

Useful? React with 👍 / 👎.

Preserve enabled state when editing routes

The TUI edit flow always sends enabled: true to update_model_route, so editing a disabled model route’s pattern/provider/priority silently re-enables it. This differs from the CLI update path, which preserves existing.enabled, and it can cause a route the user intentionally disabled to start matching requests again after a harmless edit.

Useful? React with 👍 / 👎.

Return Anthropic model fields for Claude /v1/models

When Claude Code is pointed at this proxy via ANTHROPIC_BASE_URL, its /model flow uses Anthropic's GET /v1/models schema; the Anthropic Models API documents each item with fields like type, display_name, and created_at (plus pagination fields). This handler only emits OpenAI-style entries (object, created, owned_by), so Anthropic clients that deserialize or display ModelInfo can reject the response or show no usable models even though routes exist. Consider emitting the Anthropic fields as well, or a protocol superset.

Useful? React with 👍 / 👎.

Extract Gemini route models from the request URI

For standard Gemini proxy calls, the model is in the path (for example /v1beta/models/gemini-2.5-pro:generateContent) rather than in the JSON body, so this always sets request_model to unknown for those requests. Since model-route matching now uses this value before falling back to ProviderRouter, any --app gemini model route created through the new CLI/TUI will never match normal Gemini traffic.

Useful? React with 👍 / 👎.

Preserve priority when editing model routes

When editing an existing route in the TUI, the priority prompt is always prefilled with 0, so accepting the default after changing only the pattern/provider silently rewrites any non-zero priority to highest priority. That changes routing order for existing rules even though the user did not intend to edit priority; the edit flow should prefill or preserve the route's current priority.

Useful? React with 👍 / 👎.

Run v12 compatibility repairs before returning

This compatibility branch accepts user_version=12 and returns before running the column repairs that this patch relies on for model_routes. If a v12 database already has model_routes but lacks hit_count/last_hit_at (the code comments call out that case), create_tables will not alter the existing table and later list_model_routes SELECTs those columns, causing proxy/TUI/CLI model-route operations to fail with no such column.

Useful? React with 👍 / 👎.

Repair v12 model_routes columns before returning

When opening a database with user_version 12, this compatibility branch returns after create_tables_on_conn, but CREATE TABLE IF NOT EXISTS does not alter an existing upstream model_routes table. The new DAO selects hit_count and last_hit_at, so a v12 database whose table lacks those local tracking columns will make model-route list/TUI/proxy lookups fail with no such column instead of running in compatibility mode. Run the same add_column_if_missing repairs here before returning.

Useful? React with 👍 / 👎.

Preselect the existing provider when editing

When editing an existing rule, the provider picker is always initialized at index 0. If the user only changes or accepts the pattern and presses Enter through the provider step, handle_model_route_provider_picker_key uses that selected row as the provider, so the route is silently moved to the first provider instead of preserving its current target. Initialize selected from the route's current provider_id before opening the picker.

Useful? React with 👍 / 👎.

Anchor wildcard routes that should match suffixes

For wildcard patterns that do not end in *, the generated regex is only start-anchored, so a rule such as *-4-5 also matches claude-haiku-4-55 because is_match can stop at the -4-5 prefix. Since the first matching route selects a single provider, suffix-style rules can route newer/different model IDs to the wrong provider; append $ when the pattern itself does not end with *.

Useful? React with 👍 / 👎.

Initialize edit picker to the current provider

When editing an existing route, the provider picker always starts at index 0, so pressing Enter to keep the provider silently changes the route to the first provider whenever the current provider is not first in data.providers.rows. Seed selected from the existing route's provider_id so editing pattern/priority does not rewrite the provider unintentionally.

Useful? React with 👍 / 👎.

Scope route mutations to the selected app

When a route id from another app is supplied, this deletes it solely by id even though the command is executed under the currently selected --app; the toggle and update paths use the same unscoped id-only DAO calls. Since list/add are app-scoped, a user can accidentally remove or modify a Codex route while running --app claude; verify the existing route's app_type matches app or use app-scoped DAO mutations before changing it.

Useful? React with 👍 / 👎.

Include exact route patterns in the model list

For an exact model route such as gpt-5 or deepseek-v4-pro, this match falls through to _ and adds nothing to /v1/models, so clients that populate model pickers from the proxy can miss configured route-only models or see an empty list when providers do not define env model mappings. Exact patterns without * should be emitted as model ids in addition to the role shortcuts.

Useful? React with 👍 / 👎.

Select models for the requesting app

When an OpenAI/Codex client calls the newly registered GET /v1/models endpoint, this hard-coded app scope means the response is built from Claude providers and Claude model routes only; Codex/Gemini model routes created with --app codex or --app gemini will never appear in model discovery even though request routing uses those app-specific routes. I checked server.rs, and this handler is the only /v1/models route, so the app context needs to be passed/inferred instead of fixed to Claude.

Useful? React with 👍 / 👎.

Accept the documented --provider flag

The new command docs/requirements describe model-route update <id> ... [--provider], but with #[arg(long)] on the provider_id field Clap exposes the flag as the case-converted field name (--provider-id; Clap derive reference: https://docs.rs/clap/latest/clap/_derive/index.html#arg-attributes). Users following the documented --provider form will get an unknown-argument error when trying to move a route to a different provider, so add long = "provider" or an alias.

Useful? React with 👍 / 👎.

Avoid routing model-less requests as unknown

When a request has no model field/path model, this sentinel is still passed into ModelRouter; a catch-all route such as * matches "unknown". That changes model-less endpoints like Codex /v1/responses/compact (the existing passthrough test sends only input) from using the current provider to being forced through the catch-all route, even though there is no model to route on. Keep the model as optional for routing, or skip model-route lookup when extraction fails.

Useful? React with 👍 / 👎.

Preserve the original takeover live backup

When automatic failover succeeds while proxy takeover is active, this now overwrites proxy_live_backup with the failover provider. That backup is later restored verbatim by restore_live_config_for_app when takeover is disabled or recovered, so the user's original live config captured at takeover start is lost; the hot-switch path still avoids this by writing a failover live snapshot instead. Refresh/write the failover snapshot here rather than mutating the live backup.

Useful? React with 👍 / 👎.

Restore cargo fmt compliance

Running cargo fmt --check under src-tauri/ fails on this match arm because rustfmt expects a trailing comma after the KeyCode::Enter arm body. The repository guidance notes CI runs cargo fmt --check for src-tauri/**, so this formatting miss will block the PR even though the code otherwise parses.

Useful? React with 👍 / 👎.