Skip to main content
Private AI Proxy runs on your computer. It signs in to your RedPill account, verifies the gateway’s attestation before any prompt leaves your machine, and gives your agents a local API so they never hold your RedPill credential. It is open source in Dstack-TEE/private-ai-gateway and published by dstack TEE.
Private AI Proxy overview screen, light themePrivate AI Proxy overview screen, dark theme

Overview: Protected, the current session, the Local API, and connected agents

The sections below follow the app’s screens: Profiles, Privacy, Agents, Local API, Usage, and Settings, then the CLI.

Install

Desktop app

Start building detects your platform and downloads the current stable build.
Linux requires glibc 2.35 or newer. Windows Desktop is currently Preview.

CLI

The official scripts are the recommended installation method. They detect the operating system and CPU, verify the release checksum, and install pap, private-ai-proxy, and aci.
On Homebrew 6+, each fully qualified command above automatically taps the repository and trusts only the item being installed. To use short names such as brew install private-ai-proxy or brew upgrade private-ai-proxy, trust the whole tap first:
Trust for an item installed by its fully qualified name persists, so later upgrades continue to work. Older Homebrew versions do not have tap trust; the fully qualified one-line commands still work there. Verify the installation:
Rerun an official script to update a script-managed installation. Homebrew and npm manage their own upgrades and removal. Release signatures and checksums are on the GitHub Releases page.

Profiles and sign-in

A profile is one RedPill workspace plus its credential. Add one with either:
  • Account login: the app shows a device code and opens your browser. Approve the code, then pick a workspace in the app to receive an inference key for that workspace. The flow uses device authorization (RFC 8628), with no local callback, client secret, or refresh token; the sign-in token stays in memory. If saving fails, Retry reuses the approval.
  • API key: Get API key opens the dashboard; paste the key.
Credentials live in an owner-only, plaintext credentials.toml file in the app’s settings directory. The operating system’s credential store is used only to migrate old installations; profile exports never contain credentials. Sign in again or paste a new key to replace one; revoke old keys in the dashboard. Phala Cloud profiles use the same form with Phala’s device authorization; custom endpoints accept API keys only.

Privacy: what Protected does

Switching on Protected runs the ACI relying-party checks against the gateway with a fresh nonce:
  1. the hardware quote verifies to the TEE vendor root and binds the report data;
  2. the attested keyset digest matches the statement for that nonce;
  3. the keyset has not expired;
  4. source provenance is present and the app compose matches its measured hash; and
  5. the TLS key actually used for the connection is in the attested keyset.
On success the app pins the channel to that keyset for the session and enables inference through the Local API. A verification failure or network loss keeps connected agents pointed at the Local API, which refuses requests until protection is verified again. Stopping Protected, choosing Stop All and Quit, stopping the service, or disconnecting an agent restores its previous configuration. Quitting only the desktop app leaves the backend running. Verification is never restored from disk after a restart; it is redone.

Agents

The Agents screen lists the agents the app can configure. Connect writes RedPill into the agent’s own provider settings, in the protocol the agent speaks, with an agent-scoped token for the Local API. Disconnect restores the file you had before; edits you made in the meantime are preserved, and an incomplete restore stays retryable. Only agents that are both connected and currently protected are authorized on the Local API. Connecting from the CLI previews the change first and applies it with a revision, so a file that changed in between is rejected rather than overwritten.

Local API

While Protected is on, the app serves an OpenAI-compatible API on your machine: Any tool that accepts an OpenAI-compatible endpoint can use it, for example Cline. The Local API is plain HTTP on loopback by default; send ordinary request bodies. It rejects end-to-end encryption headers because the app already owns the verified channel. Its listener runs with the backend, but inference is refused without verified protection. Rotate the client key from Local API settings or token rotate.

Usage

Responses stream to the agent immediately. The app fetches the signed receipt for each response afterwards and audits it against the attestation. Usage records every request with the agent, model, tokens, estimated cost, and the audit result; an audit failure is recorded but cannot recall bytes the agent already received. The Overview screen shows the current session’s totals. Usage can be filtered, exported to CSV, and cleared.

Settings

  • Protect on launch starts protection when the background service starts.
  • Open at Login is an operating system preference set from the desktop app.
  • Advanced > Reset settings stops protection, disconnects agents, restores defaults, generates a new web UI password, and disables Open at Login. Profiles, API keys, the local client key, and usage history are kept.
  • Closing the window or choosing Quit from the tray or menu leaves the backend running. Stop All and Quit stops protection, restores agent configuration, and stops the backend.

CLI

pap drives the same background service without a window. It comes with the desktop app or on its own. The macOS app registers it automatically; add ~/.local/bin to your terminal’s PATH if needed. The Windows installer registers it on the user PATH, and Linux packages install it system-wide. Settings > Advanced > pap command manages registration in the desktop app. private-ai-proxy and aci are the same executable.
Credentials are entered through a hidden prompt or --key-stdin, never as an argument. For scripts, --json --non-interactive returns data and --yes approves a change; agent changes are previewed with --dry-run and applied with the returned --revision. Exit status is 0 on success, 2 for invalid arguments, 1 for operation failures. verify, sessions, and serve work against any ACI service without a profile; serve refuses to start unless verification passes.

Troubleshooting