# llmx-responses-api-proxy

A strict HTTP proxy that only forwards `POST` requests to `/v1/responses` to the OpenAI API (`https://api.openai.com`), injecting the `Authorization: Bearer $OPENAI_API_KEY` header. Everything else is rejected with `403 Forbidden`.

## Expected Usage

**IMPORTANT:** `llmx-responses-api-proxy` is designed to be run by a privileged user with access to `OPENAI_API_KEY` so that an unprivileged user cannot inspect or tamper with the process. Though if `--http-shutdown` is specified, an unprivileged user _can_ make a `GET` request to `/shutdown` to shutdown the server, as an unprivileged user could not send `SIGTERM` to kill the process.

A privileged user (i.e., `root` or a user with `sudo`) who has access to `OPENAI_API_KEY` would run the following to start the server, as `llmx-responses-api-proxy` reads the auth token from `stdin`:

```shell
printenv OPENAI_API_KEY | env -u OPENAI_API_KEY llmx-responses-api-proxy --http-shutdown --server-info /tmp/server-info.json
```

A non-privileged user would then run LLMX as follows, specifying the `model_provider` dynamically:

```shell
PROXY_PORT=$(jq .port /tmp/server-info.json)
PROXY_BASE_URL="http://127.0.0.1:${PROXY_PORT}"
llmx exec -c "model_providers.openai-proxy={ name = 'OpenAI Proxy', base_url = '${PROXY_BASE_URL}/v1', wire_api='responses' }" \
    -c model_provider="openai-proxy" \
    'Your prompt here'
```

When the unprivileged user was finished, they could shutdown the server using `curl` (since `kill -SIGTERM` is not an option):

```shell
curl --fail --silent --show-error "${PROXY_BASE_URL}/shutdown"
```

## Behavior

- Reads the API key from `stdin`. All callers should pipe the key in (for example, `printenv OPENAI_API_KEY | llmx-responses-api-proxy`).
- Formats the header value as `Bearer <key>` and attempts to `mlock(2)` the memory holding that header so it is not swapped to disk.
- Listens on the provided port or an ephemeral port if `--port` is not specified.
- Accepts exactly `POST /v1/responses` (no query string). The request body is forwarded to `https://api.openai.com/v1/responses` with `Authorization: Bearer <key>` set. All original request headers (except any incoming `Authorization`) are forwarded upstream, with `Host` overridden to `api.openai.com`. For other requests, it responds with `403`.
- Optionally writes a single-line JSON file with server info, currently `{ "port": <u16>, "pid": <u32> }`.
- Optional `--http-shutdown` enables `GET /shutdown` to terminate the process with exit code `0`. This allows one user (e.g., `root`) to start the proxy and another unprivileged user on the host to shut it down.

## CLI

```
llmx-responses-api-proxy [--port <PORT>] [--server-info <FILE>] [--http-shutdown] [--upstream-url <URL>]
```

- `--port <PORT>`: Port to bind on `127.0.0.1`. If omitted, an ephemeral port is chosen.
- `--server-info <FILE>`: If set, the proxy writes a single line of JSON with `{ "port": <PORT>, "pid": <PID> }` once listening.
- `--http-shutdown`: If set, enables `GET /shutdown` to exit the process with code `0`.
- `--upstream-url <URL>`: Absolute URL to forward requests to. Defaults to `https://api.openai.com/v1/responses`.
- Authentication is fixed to `Authorization: Bearer <key>` to match the LLMX CLI expectations.

For Azure, for example (ensure your deployment accepts `Authorization: Bearer <key>`):

```shell
printenv AZURE_OPENAI_API_KEY | env -u AZURE_OPENAI_API_KEY llmx-responses-api-proxy \
  --http-shutdown \
  --server-info /tmp/server-info.json \
  --upstream-url "https://YOUR_PROJECT_NAME.openai.azure.com/openai/deployments/YOUR_DEPLOYMENT/responses?api-version=2025-04-01-preview"
```

## Notes

- Only `POST /v1/responses` is permitted. No query strings are allowed.
- All request headers are forwarded to the upstream call (aside from overriding `Authorization` and `Host`). Response status and content-type are mirrored from upstream.

## Hardening Details

Care is taken to restrict access/copying to the value of `OPENAI_API_KEY` retained in memory:

- We leverage [`llmx_process_hardening`](https://github.com/valknar/llmx/blob/main/llmx-rs/process-hardening/README.md) so `llmx-responses-api-proxy` is run with standard process-hardening techniques.
- At startup, we allocate a `1024` byte buffer on the stack and copy `"Bearer "` into the start of the buffer.
- We then read from `stdin`, copying the contents into the buffer after `"Bearer "`.
- After verifying the key matches `/^[a-zA-Z0-9_-]+$/` (and does not exceed the buffer), we create a `String` from that buffer (so the data is now on the heap).
- We zero out the stack-allocated buffer using https://crates.io/crates/zeroize so it is not optimized away by the compiler.
- We invoke `.leak()` on the `String` so we can treat its contents as a `&'static str`, as it will live for the rest of the process.
- On UNIX, we `mlock(2)` the memory backing the `&'static str`.
- When using the `&'static str` when building an HTTP request, we use `HeaderValue::from_static()` to avoid copying the `&str`.
- We also invoke `.set_sensitive(true)` on the `HeaderValue`, which in theory indicates to other parts of the HTTP stack that the header should be treated with "special care" to avoid leakage:

https://github.com/hyperium/http/blob/439d1c50d71e3be3204b6c4a1bf2255ed78e1f93/src/header/value.rs#L346-L376