# Development state

Release checkpoint: 2026-09-19. Branch: `main`. The implementation and
license baseline is commit `5767b90` (`release: finalize wgcloak v1 under
GPLv3`). The final documentation closeout is the only work that follows that
baseline and is intended to be committed separately; no implementation claim
below depends on an uncommitted code change.

## Implemented behavior

- The package provides a reusable Rust library and thin binary. Integration
  tests use the library rather than path-including production modules.
- Client and server runtimes support authenticated direct UDP, QUIC DATAGRAM,
  HTTP/3 CONNECT-UDP with HTTP Datagrams, and TLS/TCP length-framed fallback.
  TLS client connections use normal certificate-chain and hostname verification.
  Shared-secret authentication precedes encrypted backend-session activation.
- `auto` follows configured transport order, uses bounded probes/retries, stops
  on authentication rejection, and reconnects with jittered 250 ms–30 s
  backoff. Successful connections reset the backoff.
- Enabled server listeners run concurrently, reject incompatible UDP bind
  conflicts, isolate backend UDP sockets per authenticated session, enforce
  admission limits, and publish aggregate status/statistics.
- The Unix control socket uses a private runtime directory and `0600` socket
  permissions. It refuses active sockets, removes only verified stale sockets,
  parses/serializes newline-delimited JSON structurally, limits concurrent
  clients, and bounds idle/oversized requests. `status`, `stats`, `shutdown`,
  and client-only `reconnect` are supported. `reload` explicitly reports
  unsupported.
- Secret-file loading opens one descriptor, uses Linux `O_NOFOLLOW` and
  `O_NONBLOCK`, validates a regular file and permissions from that descriptor,
  and reads the opened descriptor. Secret contents are not logged.
- Systemd templates run as an unprivileged user with hardening enabled;
  `CAP_NET_BIND_SERVICE` is documented as an optional, narrowly scoped aid for
  ports below 1024. No WireGuard, TUN/TAP, routing, or firewall management is
  implemented.
- The configured default application datagram is 1,200 bytes and the project
  safety ceiling is 16,384 bytes. Transport negotiation and framing may lower
  the effective per-session maximum; boundary coverage exists in config and
  server-capability tests.
- Oversized-datagram accounting is preserved in aggregate status/statistics
  across the direct, QUIC, HTTP/3, and TLS/TCP paths; the latest boundary and
  metrics repair is implemented by `src/server/mod.rs::record_transport_drop`
  and covered by its `oversized_transport_error_is_counted_once` test.
- `tests/encrypted_runtime.rs` covers daemon-level QUIC/TLS forwarding,
  authentication rejection, cancellation, session identity, and QUIC restart
  reconnect. `tests/server_capabilities.rs` covers the datagram capability
  boundary.
- A deterministic Criterion target exists in `benches/protocol_transports.rs`
  for protocol/framing/HTTP Datagram/replay microbenchmarks. The repository
  also contains four isolated cargo-fuzz targets under `fuzz/` for direct
  packets, TLS frames, HTTP/3 parsing, and TOML parsing.
- Package metadata declares SPDX `GPL-3.0-only`, and `LICENSE` contains the
  complete GNU GPL version 3 text.

## Verification at 5767b90

The final release gates passed against the committed `5767b90` implementation
baseline before this documentation closeout:

- `cargo fmt --all -- --check`
- `cargo check --all-targets`
- `cargo clippy --all-targets -- -D warnings`
- `cargo test --tests -- --test-threads=1` — 100 tests passed
- `cargo check --manifest-path fuzz/Cargo.toml`
- `cargo bench --bench protocol_transports --no-run`

`cargo-fuzz` is not installed on this host, so no fuzzing session or sanitizer
run is claimed. No network/loopback benchmark was run; only deterministic
benchmark compilation was checked. These are deferred optional release
artifacts, not hidden implementation failures.

## Deferred optional work

- Run bounded cargo-fuzz smoke sessions and sanitizer checks when the
  `cargo-fuzz` tool is available. The targets compile and have seed corpora.
- Run an explicitly opt-in network/loopback benchmark only when its host,
  kernel, MTU, scheduler, loss, latency, and duration controls are recorded.
- Config and certificate/key reload remain unsupported. The control command
  reports this explicitly; the specification makes certificate reload
  desirable but optional and does not require full hot reload for the first
  prototype.
- Produce measured benchmark and fuzz reports if a release process requires
  them; the deterministic benchmark and compile-only fuzz evidence are
  already part of the committed release.

## Optional follow-up tasks

1. Install `cargo-fuzz` and run the four documented bounded smoke sessions.
2. Capture a reproducible deterministic benchmark report, then separately
   decide whether an environment-sensitive network benchmark is useful.
3. Revisit reload scope only if the product promotes the optional
   configuration/certificate reload behavior into a required feature.