feat: add lossless inference proxy transport

Cargo.toml

@@ -1,14 +1,23 @@
 [package]
 name = "maple-proxy"
-version = "0.1.12"
+version = "0.2.0"
 edition = "2021"
 authors = ["OpenSecret"]
 description = "Lightweight OpenAI-compatible proxy server for Maple/OpenSecret TEE infrastructure"
 license = "MIT"
-repository = "https://github.com/OpenSecret/maple-proxy"
-homepage = "https://github.com/OpenSecret/maple-proxy"
+repository = "https://github.com/OpenSecretCloud/maple-proxy"
+homepage = "https://github.com/OpenSecretCloud/maple-proxy"
 keywords = ["openai", "proxy", "tee", "opensecret", "maple"]
 categories = ["web-programming::http-server", "api-bindings"]
+include = [
+    "/src/**",
+    "/examples/**",
+    "/tests/**",
+    "/Cargo.toml",
+    "/Cargo.lock",
+    "/README.md",
+    "/LICENSE",
+]
 
 [lib]
 name = "maple_proxy"
@@ -19,13 +28,13 @@ name = "maple-proxy"
 path = "src/main.rs"
 
 [dependencies]
-# OpenSecret SDK 
-opensecret = "3.3.0"
+# OpenSecret SDK
+opensecret = "3.4.0"
 
 # Web server
 axum = { version = "0.8.4", features = ["http2", "macros"] }
 tokio = { version = "1.47", features = ["net", "rt-multi-thread", "macros", "sync", "time"] }
-tower = { version = "0.5.2", features = ["full"] }
+tower = { version = "0.5.2", features = ["util"] }
 tower-http = { version = "0.6.6", features = ["cors", "trace"] }
 
 # Serialization
@@ -51,6 +60,3 @@ http = "1.0"
 
 [dev-dependencies]
 axum-test = "18.0.1"
-tokio-test = "0.4"
-tower = { version = "0.5.2", features = ["util"] }
-serde_json = "1.0"

README.md

@@ -1,14 +1,17 @@
 # 🍁 Maple Proxy
 
-A lightweight OpenAI-compatible proxy server for Maple/OpenSecret's TEE infrastructure. Works with **any** OpenAI client library while providing the security and privacy benefits of Trusted Execution Environment (TEE) processing.
+A lightweight proxy for Maple/OpenSecret's OpenAI-compatible inference
+endpoints, with the security and privacy benefits of Trusted Execution
+Environment (TEE) processing.
 
 ## 🚀 Features
 
-- **100% OpenAI Compatible** - Drop-in replacement for OpenAI API
+- **OpenAI-Compatible Surface** - Models, chat completions, and embeddings endpoints
 - **Secure TEE Processing** - All requests processed in secure enclaves
-- **Streaming Support** - Full Server-Sent Events streaming for chat completions
+- **Lossless Chat Parameters** - Provider-specific request fields pass through unchanged
+- **Streaming and Non-Streaming** - Supports both chat completion response modes
 - **Flexible Authentication** - Environment variables or per-request API keys
-- **Zero Client Changes** - Works with existing OpenAI client code
+- **Familiar Clients** - Point compatible OpenAI clients at the proxy base URL
 - **Lightweight** - Minimal overhead, maximum performance
 - **CORS Support** - Ready for web applications
 
@@ -30,7 +33,7 @@ Add to your `Cargo.toml`:
 [dependencies]
 maple-proxy = { git = "https://github.com/opensecretcloud/maple-proxy" }
 # Or if published to crates.io:
-# maple-proxy = "0.1.0"
+# maple-proxy = "0.2.0"
 ```
 
 ## ⚙️ Configuration
@@ -40,7 +43,7 @@ Set environment variables or use command-line arguments:
 ```bash
 # Environment Variables
 export MAPLE_HOST=127.0.0.1                    # Server host (default: 127.0.0.1)
-export MAPLE_PORT=3000                         # Server port (default: 3000)
+export MAPLE_PORT=8080                         # Server port (default: 8080)
 export MAPLE_BACKEND_URL=http://localhost:3000         # Maple backend URL (prod: https://enclave.trymaple.ai)
 export MAPLE_API_KEY=your-maple-api-key        # Default API key (optional)
 export MAPLE_DEBUG=true                        # Enable debug logging
@@ -70,7 +73,7 @@ You should see:
 📋 Available endpoints:
    GET  /health              - Health check
    GET  /v1/models           - List available models
-   POST /v1/chat/completions - Create chat completions (streaming)
+   POST /v1/chat/completions - Create chat completions (streaming & non-streaming)
    POST /v1/embeddings       - Create embeddings
 ```
 
@@ -82,7 +85,7 @@ curl http://localhost:8080/v1/models \
   -H "Authorization: Bearer YOUR_MAPLE_API_KEY"
 ```
 
-#### Chat Completions (Streaming)
+#### Chat Completions
 ```bash
 curl -N http://localhost:8080/v1/chat/completions \
   -H "Authorization: Bearer YOUR_MAPLE_API_KEY" \
@@ -96,7 +99,9 @@ curl -N http://localhost:8080/v1/chat/completions \
   }'
 ```
 
-**Note:** Maple currently only supports streaming responses.
+Set `stream` to `true` for Server-Sent Events or `false` for one JSON response.
+Additional provider-specific JSON fields are forwarded without being parsed or
+rewritten by the proxy or Rust SDK.
 
 #### Embeddings
 ```bash

src/config.rs

@@ -1,5 +1,5 @@
 use clap::Parser;
-use serde::{Deserialize, Serialize};
+use serde::Serialize;
 use std::{net::SocketAddr, time::Duration};
 
 pub const DEFAULT_REQUEST_TIMEOUT_SECS: u64 = 300;
@@ -123,22 +123,22 @@ impl Config {
     }
 }
 
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct OpenAIError {
-    pub error: OpenAIErrorDetails,
+#[derive(Debug, Serialize)]
+pub(crate) struct OpenAIError {
+    error: OpenAIErrorDetails,
 }
 
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct OpenAIErrorDetails {
-    pub message: String,
+#[derive(Debug, Serialize)]
+struct OpenAIErrorDetails {
+    message: String,
     #[serde(rename = "type")]
-    pub error_type: String,
-    pub param: Option<String>,
-    pub code: Option<String>,
+    error_type: String,
+    param: Option<String>,
+    code: Option<String>,
 }
 
 impl OpenAIError {
-    pub fn new(message: impl Into<String>, error_type: impl Into<String>) -> Self {
+    fn new(message: impl Into<String>, error_type: impl Into<String>) -> Self {
         Self {
             error: OpenAIErrorDetails {
                 message: message.into(),
@@ -149,11 +149,11 @@ impl OpenAIError {
         }
     }
 
-    pub fn authentication_error(message: impl Into<String>) -> Self {
+    pub(crate) fn authentication_error(message: impl Into<String>) -> Self {
         Self::new(message, "invalid_request_error")
     }
 
-    pub fn server_error(message: impl Into<String>) -> Self {
+    pub(crate) fn server_error(message: impl Into<String>) -> Self {
         Self::new(message, "server_error")
     }
 }

src/lib.rs

@@ -1,8 +1,8 @@
-pub mod config;
-pub mod proxy;
+mod config;
+mod proxy;
 
-pub use config::{Config, OpenAIError, OpenAIErrorDetails};
-use proxy::{create_chat_completion, create_embeddings, health_check, list_models, ProxyState};
+pub use config::Config;
+use proxy::{health_check, proxy_openai_request, ProxyState};
 
 use axum::{
     extract::DefaultBodyLimit,
@@ -18,20 +18,23 @@ use tower_http::{
 };
 use tracing::Level;
 
-pub const MAX_PROXY_REQUEST_BODY_BYTES: usize = 50 * 1024 * 1024;
+const MAX_PROXY_REQUEST_BODY_BYTES: usize = 50 * 1024 * 1024;
 
 /// Create the Axum application with the given configuration
 pub fn create_app(config: Config) -> Router {
     let state = Arc::new(ProxyState::new(config.clone()));
+    create_app_with_state(config, state)
+}
 
+pub(crate) fn create_app_with_state(config: Config, state: Arc<ProxyState>) -> Router {
     let mut app = Router::new()
         // Health check endpoints
         .route("/health", get(health_check))
         .route("/", get(health_check))
         // OpenAI-compatible endpoints
-        .route("/v1/models", get(list_models))
-        .route("/v1/chat/completions", post(create_chat_completion))
-        .route("/v1/embeddings", post(create_embeddings))
+        .route("/v1/models", get(proxy_openai_request))
+        .route("/v1/chat/completions", post(proxy_openai_request))
+        .route("/v1/embeddings", post(proxy_openai_request))
         .with_state(state)
         .layer(
             ServiceBuilder::new()

src/main.rs

@@ -49,7 +49,7 @@ async fn main() -> anyhow::Result<()> {
     info!(
         "   Set MAPLE_API_KEY environment variable or provide Authorization: Bearer <key> header"
     );
-    info!("   Compatible with any OpenAI client library!");
+    info!("   OpenAI-compatible clients can use this proxy as their base URL");
     info!("");
     info!("🔗 Example curl:");
     info!("   curl http://{} \\", config.socket_addr()?);