docs: move to folder and update to new challenge payload

This commit is contained in:
Skipper
2026-04-18 15:14:03 +02:00
parent 9ab074170b
commit fd25de32a1
4 changed files with 13 additions and 153 deletions

2
.gitignore vendored
View File

@@ -3,4 +3,4 @@ scripts/__pycache__/
.DS_Store .DS_Store
.cargo/config.toml .cargo/config.toml
.vscode/ .vscode/
docs/ docs/superpowers

150
CLAUDE.md
View File

@@ -1,149 +1 @@
# CLAUDE.md Refer to @AGENTS.md for instructions.
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
## Project Overview
Arbiter is a **permissioned signing service** for cryptocurrency wallets. It consists of:
- **`server/`** — Rust gRPC daemon that holds encrypted keys and enforces policies
- **`useragent/`** — Flutter desktop app (macOS/Windows) with a Rust backend via Rinf
- **`protobufs/`** — Protocol Buffer definitions shared between server and client
The vault never exposes key material; it only produces signatures when requests satisfy configured policies.
## Toolchain Setup
Tools are managed via [mise](https://mise.jdx.dev/). Install all required tools:
```sh
mise install
```
Key versions: Rust 1.93.0 (with clippy), Flutter 3.38.9-stable, protoc 29.6, diesel_cli 2.3.6 (sqlite).
## Server (Rust workspace at `server/`)
### Crates
| Crate | Purpose |
|---|---|
| `arbiter-proto` | Generated gRPC stubs + protobuf types; compiled from `protobufs/*.proto` via `tonic-prost-build` |
| `arbiter-server` | Main daemon — actors, DB, EVM policy engine, gRPC service implementation |
| `arbiter-useragent` | Rust client library for the user agent side of the gRPC protocol |
| `arbiter-client` | Rust client library for SDK clients |
### Common Commands
```sh
cd server
# Build
cargo build
# Run the server daemon
cargo run -p arbiter-server
# Run all tests (preferred over cargo test)
cargo nextest run
# Run a single test
cargo nextest run <test_name>
# Lint
cargo clippy
# Security audit
cargo audit
# Check unused dependencies
cargo shear
# Run snapshot tests and update snapshots
cargo insta review
```
### Architecture
The server is actor-based using the **kameo** crate. All long-lived state lives in `GlobalActors`:
- **`Bootstrapper`** — Manages the one-time bootstrap token written to `~/.arbiter/bootstrap_token` on first run.
- **`Vault`** — Holds the encrypted root key and manages the Sealed/Unsealed vault state machine. On unseal, decrypts the root key into a `memsafe` hardened memory cell.
- **`FlowCoordinator`** — Coordinates cross-connection flow between user agents and SDK clients.
- **`EvmActor`** — Handles EVM transaction policy enforcement and signing.
Per-connection actors live under `actors/user_agent/` and `actors/client/`, each with `auth` (challenge-response authentication) and `session` (post-auth operations) sub-modules.
**Database:** SQLite via `diesel-async` + `bb8` connection pool. Schema managed by embedded Diesel migrations in `crates/arbiter-server/migrations/`. DB file lives at `~/.arbiter/arbiter.sqlite`. Tests use a temp-file DB via `db::create_test_pool()`.
**Cryptography:**
- Authentication: ed25519 (challenge-response, nonce-tracked per peer)
- Encryption at rest: XChaCha20-Poly1305 (versioned via `scheme` field for transparent migration on unseal)
- Password KDF: Argon2
- Unseal transport: X25519 ephemeral key exchange
- TLS: self-signed certificate (aws-lc-rs backend), fingerprint distributed via `ArbiterUrl`
**Protocol:** gRPC with Protocol Buffers. The `ArbiterUrl` type encodes host, port, CA cert, and bootstrap token into a single shareable string (printed to console on first run).
### Proto Regeneration
When `.proto` files in `protobufs/` change, rebuild to regenerate:
```sh
cd server && cargo build -p arbiter-proto
```
### Database Migrations
```sh
# Create a new migration
diesel migration generate <name> --migration-dir crates/arbiter-server/migrations
# Run migrations manually (server also runs them on startup)
diesel migration run --migration-dir crates/arbiter-server/migrations
```
### Code Conventions
**`#[must_use]` Attribute:**
Apply the `#[must_use]` attribute to return types of functions where the return value is critical and should not be accidentally ignored. This is commonly used for:
- Methods that return `bool` indicating success/failure or validation state
- Any function where ignoring the return value indicates a logic error
Do not apply `#[must_use]` redundantly to items (types or functions) that are already annotated with `#[must_use]`.
Example:
```rust
#[must_use]
pub fn verify(&self, nonce: i32, context: &[u8], signature: &Signature) -> bool {
// verification logic
}
```
This forces callers to either use the return value or explicitly ignore it with `let _ = ...;`, preventing silent failures.
## User Agent (Flutter + Rinf at `useragent/`)
The Flutter app uses [Rinf](https://rinf.cunarist.org) to call Rust code. The Rust logic lives in `useragent/native/hub/` as a separate crate that uses `arbiter-useragent` for the gRPC client.
Communication between Dart and Rust uses typed **signals** defined in `useragent/native/hub/src/signals/`. After modifying signal structs, regenerate Dart bindings:
```sh
cd useragent && rinf gen
```
### Common Commands
```sh
cd useragent
# Run the app (macOS or Windows)
flutter run
# Regenerate Rust↔Dart signal bindings
rinf gen
# Analyze Dart code
flutter analyze
```
The Rinf Rust entry point is `useragent/native/hub/src/lib.rs`. It spawns actors defined in `useragent/native/hub/src/actors/` which handle Dart↔server communication via signals.

View File

@@ -22,12 +22,14 @@ Arbiter distinguishes two kinds of peers:
All peers authenticate via public-key cryptography using a challenge-response protocol: All peers authenticate via public-key cryptography using a challenge-response protocol:
1. The peer sends its public key and requests a challenge. 1. The peer sends its public key and requests a challenge.
2. The server looks up the key in its database. If found, it increments the nonce and returns a challenge (replay-attack protection). 2. The server looks up the key in its database. If found, it generates a fresh challenge from random bytes plus the current timestamp.
3. The peer signs the challenge with its private key and sends the signature back. 3. The peer signs the canonical challenge payload with its private key and sends the signature back.
4. The server verifies the signature: 4. The server verifies the signature:
- **Pass:** The connection is considered authenticated. - **Pass:** The connection is considered authenticated.
- **Fail:** The server closes the connection. - **Fail:** The server closes the connection.
Authentication challenges are per-connection, ephemeral values. They are not persisted in the peer tables, and peer records store no challenge state.
### 2.2 User Agent Bootstrap ### 2.2 User Agent Bootstrap
On first run — when no User Agents are registered — the server generates a one-time bootstrap token. It is made available in two ways: On first run — when no User Agents are registered — the server generates a one-time bootstrap token. It is made available in two ways:

View File

@@ -45,7 +45,13 @@ flowchart TD
K -- yes --> J([Session started]) K -- yes --> J([Session started])
``` ```
Auth challenges are generated from fresh random bytes plus a timestamp. They are signed as the canonical challenge payload and are not persisted in `program_client`. Auth challenges are generated from fresh random bytes plus a nanosecond timestamp. The server keeps the issued challenge only in the in-flight authentication state for that connection, then verifies the signature against the same canonical challenge payload.
The authentication schema stores peer identity, not replay counters:
- `program_client` stores the SDK client's public key, metadata binding, and timestamps.
- `useragent_client` stores the User Agent public key and timestamps.
- Neither table stores an authentication nonce, and challenge generation does not update either table.
--- ---