test: add Phase 1 integration tests for basic proxy

6 tests proving the scaffold works end-to-end:
- chat completion proxied through gateway to mock backend
- /health endpoint with healthy node
- /v1/models returns seeded model list
- 404 for unknown model
- 404 when no healthy nodes available
- 400 when request body missing model field

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

crates/cortex-gateway/Cargo.toml

@@ -23,3 +23,6 @@ futures.workspace = true
 tokio-stream.workspace = true
 eventsource-stream.workspace = true
 bytes = "1"
+
+[dev-dependencies]
+tokio = { workspace = true, features = ["test-util"] }

crates/cortex-gateway/tests/common/mod.rs

@@ -0,0 +1,115 @@
+use axum::routing::{get, post};
+use axum::{Json, Router};
+use cortex_core::config::{
+    EvictionSettings, EvictionStrategy, GatewayConfig, GatewaySettings, NodeConfig,
+};
+use cortex_core::node::{ModelEntry, ModelStatus};
+use cortex_gateway::state::CortexState;
+use serde_json::{Value, json};
+use std::sync::Arc;
+use tokio::net::TcpListener;
+
+/// Spawns a mock mistral.rs backend on a random port.
+/// Returns the base URL (e.g. "http://127.0.0.1:12345").
+pub async fn spawn_mock_backend() -> String {
+    let app = Router::new()
+        .route("/v1/chat/completions", post(mock_chat_completions))
+        .route("/v1/models", get(mock_list_models));
+
+    let listener = TcpListener::bind("127.0.0.1:0").await.unwrap();
+    let addr = listener.local_addr().unwrap();
+    tokio::spawn(async move {
+        axum::serve(listener, app).await.unwrap();
+    });
+
+    format!("http://{addr}")
+}
+
+async fn mock_chat_completions(Json(body): Json<Value>) -> Json<Value> {
+    let model = body
+        .get("model")
+        .and_then(|v| v.as_str())
+        .unwrap_or("unknown");
+
+    Json(json!({
+        "id": "chatcmpl-test-001",
+        "object": "chat.completion",
+        "created": 1700000000_u64,
+        "model": model,
+        "choices": [{
+            "index": 0,
+            "message": {
+                "role": "assistant",
+                "content": "Hello from mock backend"
+            },
+            "finish_reason": "stop"
+        }],
+        "usage": {
+            "prompt_tokens": 10,
+            "completion_tokens": 5,
+            "total_tokens": 15
+        }
+    }))
+}
+
+async fn mock_list_models() -> Json<Value> {
+    Json(json!({
+        "object": "list",
+        "data": [{
+            "id": "test-model",
+            "object": "model",
+            "status": "loaded"
+        }]
+    }))
+}
+
+/// Spawns the cortex gateway with a single node pointing at `mock_url`.
+/// The node is pre-seeded as healthy with one loaded model ("test-model").
+/// Returns the gateway's base URL.
+pub async fn spawn_gateway(mock_url: &str) -> String {
+    let config = GatewayConfig {
+        gateway: GatewaySettings {
+            listen: "127.0.0.1:0".into(),
+            metrics_listen: "127.0.0.1:0".into(),
+        },
+        eviction: EvictionSettings {
+            strategy: EvictionStrategy::Lru,
+            defrag_after_cycles: 0,
+        },
+        nodes: vec![NodeConfig {
+            name: "mock-node".into(),
+            endpoint: mock_url.to_string(),
+            vram_mb: 24000,
+            pinned: vec![],
+        }],
+    };
+
+    let fleet = Arc::new(CortexState::from_config(&config));
+
+    // Seed the node as healthy with a loaded model.
+    // (Bypasses the poller, which is not running in tests.)
+    {
+        let mut nodes = fleet.nodes.write().await;
+        let node = nodes.get_mut("mock-node").expect("node must exist");
+        node.healthy = true;
+        node.models.insert(
+            "test-model".into(),
+            ModelEntry {
+                id: "test-model".into(),
+                status: ModelStatus::Loaded,
+                last_accessed: None,
+                vram_estimate_mb: Some(8000),
+            },
+        );
+    }
+
+    let app = cortex_gateway::build_app(fleet);
+
+    let listener = TcpListener::bind("127.0.0.1:0").await.unwrap();
+    let addr = listener.local_addr().unwrap();
+    tokio::spawn(async move {
+        axum::serve(listener, app).await.unwrap();
+    });
+
+    format!("http://{addr}")
+}

crates/cortex-gateway/tests/proxy_basic.rs

@@ -0,0 +1,174 @@
+mod common;
+
+use serde_json::json;
+
+#[tokio::test]
+async fn test_chat_completion_proxy() {
+    let mock_url = common::spawn_mock_backend().await;
+    let gw_url = common::spawn_gateway(&mock_url).await;
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .post(format!("{gw_url}/v1/chat/completions"))
+        .header("content-type", "application/json")
+        .json(&json!({
+            "model": "test-model",
+            "messages": [{"role": "user", "content": "Hi"}]
+        }))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 200);
+
+    let body: serde_json::Value = resp.json().await.expect("valid JSON response");
+    assert_eq!(body["id"], "chatcmpl-test-001");
+    assert_eq!(body["model"], "test-model");
+    assert_eq!(
+        body["choices"][0]["message"]["content"],
+        "Hello from mock backend"
+    );
+    assert_eq!(body["usage"]["total_tokens"], 15);
+}
+
+#[tokio::test]
+async fn test_health_endpoint() {
+    let mock_url = common::spawn_mock_backend().await;
+    let gw_url = common::spawn_gateway(&mock_url).await;
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .get(format!("{gw_url}/health"))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 200);
+
+    let body: serde_json::Value = resp.json().await.unwrap();
+    assert_eq!(body["status"], "ok");
+    assert_eq!(body["nodes"]["healthy"], 1);
+    assert_eq!(body["nodes"]["total"], 1);
+}
+
+#[tokio::test]
+async fn test_list_models() {
+    let mock_url = common::spawn_mock_backend().await;
+    let gw_url = common::spawn_gateway(&mock_url).await;
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .get(format!("{gw_url}/v1/models"))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 200);
+
+    let body: serde_json::Value = resp.json().await.unwrap();
+    assert_eq!(body["object"], "list");
+
+    let data = body["data"].as_array().expect("data should be an array");
+    assert_eq!(data.len(), 1);
+    assert_eq!(data[0]["id"], "test-model");
+}
+
+#[tokio::test]
+async fn test_model_not_found() {
+    let mock_url = common::spawn_mock_backend().await;
+    let gw_url = common::spawn_gateway(&mock_url).await;
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .post(format!("{gw_url}/v1/chat/completions"))
+        .header("content-type", "application/json")
+        .json(&json!({
+            "model": "nonexistent-model",
+            "messages": [{"role": "user", "content": "Hi"}]
+        }))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 404);
+
+    let body: serde_json::Value = resp.json().await.unwrap();
+    assert!(
+        body["error"]["message"]
+            .as_str()
+            .unwrap()
+            .contains("not found")
+    );
+}
+
+#[tokio::test]
+async fn test_no_healthy_nodes() {
+    let config = cortex_core::config::GatewayConfig {
+        gateway: cortex_core::config::GatewaySettings {
+            listen: "127.0.0.1:0".into(),
+            metrics_listen: "127.0.0.1:0".into(),
+        },
+        eviction: cortex_core::config::EvictionSettings {
+            strategy: cortex_core::config::EvictionStrategy::Lru,
+            defrag_after_cycles: 0,
+        },
+        nodes: vec![cortex_core::config::NodeConfig {
+            name: "dead-node".into(),
+            endpoint: "http://127.0.0.1:1".into(),
+            vram_mb: 24000,
+            pinned: vec![],
+        }],
+    };
+    let fleet = std::sync::Arc::new(cortex_gateway::state::CortexState::from_config(&config));
+
+    let app = cortex_gateway::build_app(fleet);
+    let listener = tokio::net::TcpListener::bind("127.0.0.1:0").await.unwrap();
+    let addr = listener.local_addr().unwrap();
+    tokio::spawn(async move {
+        axum::serve(listener, app).await.unwrap();
+    });
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .post(format!("http://{addr}/v1/chat/completions"))
+        .header("content-type", "application/json")
+        .json(&json!({
+            "model": "any-model",
+            "messages": [{"role": "user", "content": "Hi"}]
+        }))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 404);
+
+    let body: serde_json::Value = resp.json().await.unwrap();
+    assert!(
+        body["error"]["message"]
+            .as_str()
+            .unwrap()
+            .contains("no healthy nodes")
+    );
+}
+
+#[tokio::test]
+async fn test_missing_model_field() {
+    let mock_url = common::spawn_mock_backend().await;
+    let gw_url = common::spawn_gateway(&mock_url).await;
+
+    let client = reqwest::Client::new();
+    let resp = client
+        .post(format!("{gw_url}/v1/chat/completions"))
+        .header("content-type", "application/json")
+        .json(&json!({
+            "messages": [{"role": "user", "content": "Hi"}]
+        }))
+        .send()
+        .await
+        .expect("request should succeed");
+
+    assert_eq!(resp.status(), 400);
+
+    let body: serde_json::Value = resp.json().await.unwrap();
+    assert!(body["error"]["message"].as_str().unwrap().contains("model"));
+}