refactor: unify authentication state and improve middleware efficiency

- Introduce AuthInfo struct for shared auth state.
- Populate AuthInfo in rate_limit_middleware and store in request extensions.
- Update AuthenticatedClient extractor to use pre-resolved AuthInfo.
- Simplify chat_completions by removing redundant DB lookups.

src/auth/mod.rs

@@ -1,33 +1,40 @@
 use axum::{extract::FromRequestParts, http::request::Parts};
-use axum_extra::TypedHeader;
-use axum_extra::headers::Authorization;
-use headers::authorization::Bearer;
 
 use crate::errors::AppError;
 
-pub struct AuthenticatedClient {
+#[derive(Debug, Clone)]
+pub struct AuthInfo {
     pub token: String,
     pub client_id: String,
 }
 
+pub struct AuthenticatedClient {
+    pub info: AuthInfo,
+}
+
 impl<S> FromRequestParts<S> for AuthenticatedClient
 where
     S: Send + Sync,
 {
     type Rejection = AppError;
 
-    async fn from_request_parts(parts: &mut Parts, state: &S) -> Result<Self, Self::Rejection> {
+    async fn from_request_parts(parts: &mut Parts, _state: &S) -> Result<Self, Self::Rejection> {
-        // Extract bearer token from Authorization header
+        // Retrieve AuthInfo from request extensions, where it was placed by rate_limit_middleware
-        let TypedHeader(Authorization(bearer)) = TypedHeader::<Authorization<Bearer>>::from_request_parts(parts, state)
+        let info = parts
-            .await
+            .extensions
-            .map_err(|_| AppError::AuthError("Missing or invalid bearer token".to_string()))?;
+            .get::<AuthInfo>()
+            .cloned()
+            .ok_or_else(|| AppError::AuthError("Authentication info not found in request".to_string()))?;
 
-        let token = bearer.token().to_string();
+        Ok(AuthenticatedClient { info })
+    }
+}
 
-        // Derive client_id from the token prefix
+impl std::ops::Deref for AuthenticatedClient {
-        let client_id = format!("client_{}", &token[..8.min(token.len())]);
+    type Target = AuthInfo;
 
-        Ok(AuthenticatedClient { token, client_id })
+    fn deref(&self) -> &Self::Target {
+        &self.info
     }
 }
 

src/rate_limiting/mod.rs

@@ -299,6 +299,7 @@ pub mod middleware {
     use super::*;
     use crate::errors::AppError;
     use crate::state::AppState;
+    use crate::auth::AuthInfo;
     use axum::{
         extract::{Request, State},
         middleware::Next,
@@ -309,20 +310,24 @@ pub mod middleware {
     /// Rate limiting middleware
     pub async fn rate_limit_middleware(
         State(state): State<AppState>,
-        request: Request,
+        mut request: Request,
         next: Next,
     ) -> Result<Response, AppError> {
         // Extract token synchronously from headers (avoids holding &Request across await)
         let token = extract_bearer_token(&request);
 
-        // Resolve client_id: DB token lookup, then prefix fallback
+        // Resolve client_id and populate AuthInfo: DB token lookup, then prefix fallback
-        let client_id = resolve_client_id(token, &state).await;
+        let auth_info = resolve_auth_info(token, &state).await;
+        let client_id = auth_info.client_id.clone();
 
         // Check rate limits
         if !state.rate_limit_manager.check_client_request(&client_id).await? {
             return Err(AppError::RateLimitError("Rate limit exceeded".to_string()));
         }
 
+        // Store AuthInfo in request extensions for extractors and downstream handlers
+        request.extensions_mut().insert(auth_info);
+
         Ok(next.run(request).await)
     }
 
@@ -334,26 +339,39 @@ pub mod middleware {
             .map(|t| t.to_string())
     }
 
-    /// Resolve client ID: try DB token first, then fall back to token-prefix derivation
+    /// Resolve auth info: try DB token first, then fall back to token-prefix derivation
-    async fn resolve_client_id(token: Option<String>, state: &AppState) -> String {
+    async fn resolve_auth_info(token: Option<String>, state: &AppState) -> AuthInfo {
         if let Some(token) = token {
             // Try DB token lookup first
-            if let Ok(Some(cid)) = sqlx::query_scalar::<_, String>(
+            match sqlx::query_scalar::<_, String>(
-                "SELECT client_id FROM client_tokens WHERE token = ? AND is_active = TRUE",
+                "UPDATE client_tokens SET last_used_at = CURRENT_TIMESTAMP WHERE token = ? AND is_active = TRUE RETURNING client_id",
             )
             .bind(&token)
             .fetch_optional(&state.db_pool)
             .await
             {
-                return cid;
+                Ok(Some(cid)) => {
+                    return AuthInfo {
+                        token,
+                        client_id: cid,
+                    };
+                }
+                Err(e) => {
+                    warn!("DB error during token lookup: {}", e);
+                }
+                _ => {}
             }
 
             // Fallback to token-prefix derivation (env tokens / permissive mode)
-            return format!("client_{}", &token[..8.min(token.len())]);
+            let client_id = format!("client_{}", &token[..8.min(token.len())]);
+            return AuthInfo { token, client_id };
         }
 
         // No token — anonymous
-        "anonymous".to_string()
+        AuthInfo {
+            token: String::new(),
+            client_id: "anonymous".to_string(),
+        }
     }
 
     /// Circuit breaker middleware for provider requests

src/server/mod.rs

@@ -122,32 +122,16 @@ async fn chat_completions(
     auth: AuthenticatedClient,
     Json(mut request): Json<ChatCompletionRequest>,
 ) -> Result<axum::response::Response, AppError> {
-    // Resolve client_id: try DB token first, then env tokens, then permissive fallback
+    let client_id = auth.client_id.clone();
-    let db_client_id: Option<String> = sqlx::query_scalar::<_, String>(
-        "SELECT client_id FROM client_tokens WHERE token = ? AND is_active = TRUE",
-    )
-    .bind(&auth.token)
-    .fetch_optional(&state.db_pool)
-    .await
-    .unwrap_or(None);
-
-    let client_id = if let Some(cid) = db_client_id {
-        // Update last_used_at in background (fire-and-forget)
-        let pool = state.db_pool.clone();
     let token = auth.token.clone();
-        tokio::spawn(async move {
+
-            let _ = sqlx::query("UPDATE client_tokens SET last_used_at = CURRENT_TIMESTAMP WHERE token = ?")
+    // Verify token if env tokens are configured
-                .bind(&token)
+    if !state.auth_tokens.is_empty() && !state.auth_tokens.contains(&token) {
-                .execute(&pool)
+        // If not in env tokens, check if it was a DB token (client_id wouldn't be client_XXXX prefix)
-                .await;
+        if client_id.starts_with("client_") {
-        });
-        cid
-    } else if state.auth_tokens.is_empty() || state.auth_tokens.contains(&auth.token) {
-        // Env token match or permissive mode (no env tokens configured)
-        auth.client_id.clone()
-    } else {
              return Err(AppError::AuthError("Invalid authentication token".to_string()));
-    };
+        }
+    }
 
     let start_time = std::time::Instant::now();
     let model = request.model.clone();