Commit Graph

140 Commits

Author SHA1 Message Date
511fc7822e feat(sdk): add transcript archive export and verification
SDK-level export_transcript() writes all conversation messages to an
encrypted, tamper-evident archive using the existing core transcript
format (Argon2id + ChaCha20-Poly1305, CBOR records, SHA-256 chain).
verify_transcript() supports both full decryption + chain check and
structural-only validation without the password.
2026-03-04 20:59:03 +01:00
f57dda3f36 feat(sdk): add Swift and Kotlin mobile client foundations with push token proto
Swift SDK: Swift Package wrapping libquicproquo_ffi with QpqClient class
(connect, login, send, receive, disconnect) for iOS 15+ / macOS 13+.

Kotlin SDK: JNI bridge to libquicproquo_ffi with QpqClient class for
Android (aarch64, armv7) and JVM, Gradle build configuration.

Adds RegisterPushToken RPC (method ID 710) to device.proto for
APNs/FCM/WebPush device push token registration.
2026-03-04 20:58:23 +01:00
cbb76af6b1 docs(sdk): add comprehensive SDK documentation and wire format reference
Covers all official SDKs (Rust, Go, Python, TypeScript, C FFI),
the v2 wire format with method ID tables, authentication flow,
and a build-your-own-SDK guide with implementation checklist.
2026-03-04 20:55:24 +01:00
2d56824834 feat(web): add browser web client with IndexedDB, Service Worker, and PWA
Create sdks/web/ with a vanilla TypeScript web client:
- IndexedDB local store for conversations, messages, and identity
- WebSocket transport connecting to the server bridge
- Service Worker with cache-first strategy for offline support
- PWA manifest for installable web app
- Dark-themed responsive UI with sidebar, messages, and input bar
- Connection status badge and MLS epoch indicator in header
- Unread message count badges on conversations
2026-03-04 20:55:05 +01:00
496f83067a feat(delivery): add server-signed delivery proof on enqueue
The server now produces a 96-byte Ed25519-signed delivery proof for
every enqueued message: SHA-256(seq || recipient_key || timestamp_ms)
followed by the server's Ed25519 signature. Clients can verify the
proof using verify_delivery_proof() in quicproquo-core to get
cryptographic evidence the server accepted their message.
2026-03-04 20:54:55 +01:00
1768f85258 feat(kt): add key revocation and Merkle-log audit support
Add RevocationLog with domain-separated leaf hashes (0x02 prefix) for
tracking revoked identity keys alongside the KT MerkleLog. Includes:

- RevocationLog with O(1) lookup, serialization, and double-revoke guard
- MerkleLog.append_raw() for pre-computed hashes
- MerkleLog.audit_log(start, end) for paginated log retrieval
- RevokeKey (510), CheckRevocation (511), AuditKeyTransparency (520) RPCs
- Server domain logic + v2 handlers + FileBackedStore/SqlStore persistence
- 4 new revocation tests + all 21 KT tests + 65 server tests passing
2026-03-04 20:53:41 +01:00
f667281831 feat(tui): add MLS epoch indicator, online/offline status, and 9 rendering tests
Enhance v2 TUI with connected/mls_epoch state fields, colored connection
indicator in status bar, MLS epoch display, and wildcard match for new
SDK event variants. Add 9 tests using ratatui TestBackend covering
rendering, navigation, scroll bounds, status bar content, and unread
count display. Also fix rand 0.8 compat issue in v2_repl.rs.
2026-03-04 20:52:27 +01:00
372dd67a3b feat: add OpenWrt cross-compilation and packaging (Phase F7)
- packaging/openwrt/: opkg Makefile, procd init script, uci config
- scripts/cross-compile.sh: build for musl targets with size checks
- .github/workflows/openwrt.yml: CI cross-compile + 5 MB size gate
- docs/openwrt.md: installation and configuration guide
- Targets: x86_64-musl, armv7-musleabihf, aarch64-musl
- Uses cargo-zigbuild for Docker-free cross-compilation
2026-03-04 20:52:15 +01:00
49e8e066d7 feat(sdk): add Python SDK with QUIC and FFI transport backends
Implements quicproquo-py with two transport backends:
- Async QUIC transport via aioquic with v2 protobuf wire format
- Synchronous Rust FFI transport via CFFI wrapping libquicproquo_ffi

Includes manual protobuf encode/decode (no codegen), full RPC coverage
(auth, delivery, channels, users, keys, health), PyPI-ready packaging,
async echo bot and FFI demo examples, and 15 passing unit tests.
2026-03-04 20:52:02 +01:00
f4621b3425 feat: add traffic analysis resistance (Phase 7.7 + F8)
- Core: add pad_uniform/unpad_uniform for configurable boundary padding
  and generate_decoy for indistinguishable decoy messages
- Server: add traffic_resistance module with payload padding, timing
  jitter, and background decoy traffic generator (feature-gated)
- P2P: add mesh traffic_resistance module with padded envelopes and
  mesh decoy injection (feature-gated)
- All gated behind --features traffic-resistance
- 22 new tests across core (8), server (4), and P2P (5)
2026-03-04 20:50:19 +01:00
c401caec60 feat(bench): add safety number & epoch rotation benchmarks, CI workflow
Add safety_number benchmark to crypto_benchmarks.rs, epoch rotation
(propose_self_update + commit) benchmark to mls_operations.rs, expand
add_member group sizes to include 100, and add .github/workflows/bench.yml
that runs Criterion benchmarks and uploads HTML reports as artifacts.
2026-03-04 20:49:42 +01:00
885cce0d7d feat: add multi-node horizontal scaling foundations
NotificationBus for cross-node message delivery fan-out:
- NotificationBus trait: publish(topic) + subscribe(topic) -> Notify
- InMemoryNotificationBus: single-node default via tokio::sync::Notify
- Designed for Redis pub/sub or NATS replacement in multi-node deploys
- 3 async tests: publish wakes, timeout without publish, independent topics

Health endpoint enhancements for load balancer awareness:
- HealthResponse proto: add node_id, version, uptime_secs, storage_backend
- ServerState: add node_id, start_time, storage_backend fields
- Health handler returns full node identity for multi-node monitoring
2026-03-04 20:38:59 +01:00
913f6faaf3 feat: add distributed rate limiting with sliding window algorithm
- RateLimiter trait with check_rate(key, config) -> RateResult
- InMemoryRateLimiter: DashMap-based sliding window log per key
- RateLimitConfig: configurable max_requests and window duration
- RateResult: allowed/remaining/retry_after_secs for Retry-After headers
- Lazy GC of expired entries (every 60s)
- Thread-safe concurrent access via DashMap
- 5 unit tests: limit enforcement, independent keys, remaining counter, concurrency
2026-03-04 20:35:45 +01:00
e93a38243f feat: add graceful shutdown with drain timeout and per-RPC timeouts
Graceful shutdown (Phase 6.4):
- Listen for SIGTERM + SIGINT via tokio::signal
- Configurable drain timeout (--drain-timeout / QPQ_DRAIN_TIMEOUT, default 30s)
- Health endpoint returns "draining" during shutdown for load balancer awareness
- ServerState carries atomic draining flag
- Add RpcStatus::Unavailable (9) for shutdown-related rejections

Per-RPC timeouts (Phase 6.5):
- Add RpcStatus::DeadlineExceeded (8) for server-side timeouts
- MethodRegistry supports default_timeout and per-method timeout overrides
- RPC dispatch wraps handler invocation with tokio::time::timeout
- RequestContext carries optional deadline (Instant) for handlers
- Health: 5s timeout, blob upload/download: 120s timeout, default: 30s
- Config: --rpc-timeout / QPQ_RPC_TIMEOUT, --storage-timeout / QPQ_STORAGE_TIMEOUT
2026-03-04 20:33:26 +01:00
91c5495ab7 docs: add operational runbook, Grafana dashboard, and production docker-compose
Add comprehensive operational documentation:
- docs/operations/backup-restore.md: SQLCipher, file backend, blob backup/restore
- docs/operations/key-rotation.md: auth token, TLS, federation, DB key, OPAQUE rotation
- docs/operations/incident-response.md: playbook for common incidents
- docs/operations/scaling-guide.md: resource sizing, scaling triggers, capacity planning
- docs/operations/monitoring.md: Prometheus metrics, alert rules, log monitoring
- docs/operations/dashboards/qpq-overview.json: Grafana dashboard template
- docs/operations/prometheus.yml + alerts: Prometheus scrape and alert config
- docs/operations/grafana-provisioning/: auto-provisioning for datasources and dashboards
- docker-compose.prod.yml: production stack (server + Prometheus + Grafana)
- .env.example: documented environment variable template
2026-03-04 20:30:57 +01:00
b94248b3b6 feat: implement MLS lifecycle and multi-device support
Phase 5.3 (MLS lifecycle):
- Add group.proto with RemoveMember, UpdateGroupMetadata, ListGroupMembers, RotateKeys RPCs
- Add GroupService domain logic with metadata and membership persistence
- Add v2 RPC handlers for all 4 group management endpoints (method IDs 410-413)
- Add SDK functions: remove_member_from_group, leave_group, rotate_group_keys, set_group_metadata, get_group_members
- Add REPL commands: /group remove, /group rename, /group rotate-keys, /group leave
- Add 5 unit tests for GroupService (metadata CRUD, membership add/list/remove)

Phase 5.1 (multi-device):
- Wire device_id through SDK fetch/ack functions (fetch_for_device, ack)
- Add /devices list|add|remove REPL commands with tab completion
- Add clear_failed_outbox to ConversationStore
- Fix missing message_id/device_id fields in SDK proto struct initializers
2026-03-04 20:20:55 +01:00
a90020fe89 fix(e2e): serialize all init_auth tests behind AUTH_LOCK to prevent race
Every test that calls init_auth() now holds AUTH_LOCK for its full
duration, preventing the global AUTH_CONTEXT from being overwritten
by concurrent tests. The e2e_auth_failure_wrong_token test additionally
resets auth back to "devtoken" after its assertion. Tests now pass
reliably with default parallelism (no --test-threads 1 required).
2026-03-04 20:20:03 +01:00
fd1accc6dd feat(sdk): wire device_id through messaging and client APIs
Add device_id parameter to fetch, fetch_wait, ack, receive_messages,
and receive_messages_wait SDK functions. QpqClient gains device_id
field with register_device/list_devices/revoke_device convenience
methods. Client REPL passes empty device_id for backwards compat.
2026-03-04 20:19:30 +01:00
799aab68fe feat(server): wire device_id through delivery proto and v2 handlers
Add device_id field to FetchRequest, FetchWaitRequest, PeekRequest,
and AckRequest proto messages. V2 handlers now build composite
queue keys (identity_key + device_id) when device_id is provided,
enabling per-device fetch/ack scoping.
2026-03-04 20:16:41 +01:00
eaca24397b feat(server): add multi-device delivery fan-out
Enqueue now resolves all registered devices for a recipient identity
and fans out the message to each device-scoped queue. Single-device
clients remain backwards compatible (bare identity_key queue).

Also adds FileBackedStore::ephemeral() test helper.
2026-03-04 20:15:26 +01:00
12b19b6931 feat: implement account recovery with encrypted backup bundles
Add recovery code generation (8 codes per setup), Argon2id key derivation,
ChaCha20-Poly1305 encrypted bundles, and server-side zero-knowledge storage.
Each code independently recovers the account. Includes core crypto module,
protobuf service (method IDs 750-752), server domain + handlers, SDK methods,
SQL migration, and CLI commands (/recovery setup, /recovery restore).
2026-03-04 20:12:20 +01:00
5b6d8209f0 feat: add abuse prevention and moderation (Phase 5.6)
Add server-side moderation service with report submission, user
banning/unbanning, and admin listing endpoints. Add client-side
user blocking with message filtering in ConversationStore.

Server:
- ModerationService domain logic (report, ban, unban, list)
- Storage trait methods + FileBackedStore + SqlStore implementations
- SQL migration 012_moderation.sql (reports + bans tables)
- Error codes E031-E033 for moderation
- Domain types for all moderation request/response pairs
- 10 new tests (6 domain + 4 storage)

SDK:
- blocked_users table in ConversationStore
- block_user, unblock_user, is_blocked, list_blocked methods
- load_recent_messages_filtered excludes blocked senders
- QpqClient moderation convenience methods
- 4 new tests for block/unblock/filter
2026-03-04 20:11:20 +01:00
a1f0dbc514 docs: mark Phases 1, 2, and 4.4 complete in ROADMAP
Phase 1 (production hardening): all 5 items done
Phase 2 (test & CI maturity): all 4 items done
Phase 4.4 (PQ-MLS): already implemented — hybrid KEM in OpenMLS provider
2026-03-04 15:28:03 +01:00
5a66c2e954 chore: fix all clippy warnings across workspace 2026-03-04 14:13:58 +01:00
4013b223ff test(e2e): add auth failure, message ordering, OPAQUE flow, key exhaustion, and rate limit tests 2026-03-04 13:33:21 +01:00
3a42130518 chore(ci): enforce clippy -D warnings, add v2 branch trigger 2026-03-04 13:31:32 +01:00
c8c5f96ecd chore: uncomment CODEOWNERS entries with crate-specific paths 2026-03-04 13:31:29 +01:00
e5329ee8e5 test: add unit tests for ConversationStore CRUD, outbox, and ConversationId 2026-03-04 13:31:25 +01:00
e3dfc43e2c test: add unit tests for token cache round-trip and edge cases 2026-03-04 13:31:22 +01:00
7bcfbf175c test: add unit tests for v2 REPL split_cmd parsing 2026-03-04 13:31:19 +01:00
75f11cb76b test: add unit tests for retry logic and retriable classifier 2026-03-04 13:31:16 +01:00
a3f67aca45 feat(tls): add certificate expiry validation and self-signed warning 2026-03-04 13:30:46 +01:00
950f477842 fix(docker): sync Dockerfile with v2 workspace, add production env and volume docs 2026-03-04 13:30:20 +01:00
3393514911 fix: tighten production config validation with token length and db path checks 2026-03-04 13:29:50 +01:00
a8ed3c4356 fix: replace unwrap/expect in production paths with fallible alternatives 2026-03-04 13:29:33 +01:00
cab03bd3f7 feat(client): v2 REPL over SDK with categorized help and tab-completion
925-line REPL replacing the 3317-line monolith — delegates all crypto,
MLS, and RPC to quicproquo-sdk. 20 commands across 6 categories
(messaging, groups, account, keys, utility, debug), rustyline tab
completion, background event listener, auto-server-launch.

Also adds SDK accessor methods (server_addr_string, config_state_path),
WS bridge register handler, and README table formatting cleanup.
2026-03-04 13:02:54 +01:00
99f9abe9ed feat(client): ratatui TUI subscribing to SDK events
Single-file v2 TUI (v2_tui.rs, ~580 lines) gated behind
cfg(all(feature = "v2", feature = "tui")).

Layout: conversations sidebar, scrollable messages panel,
input bar with cursor editing, status line. Help overlay
on /help. Subscribes to SDK broadcast::Receiver<ClientEvent>
for real-time updates. TerminalGuard ensures terminal restore
on panic. Commands: /login, /register, /logout, /dm, /group,
/quit, /help.

Build: cargo build -p quicproquo-client --features v2,tui
2026-03-04 12:47:15 +01:00
029c701780 feat(client): v2 CLI binary over SDK with simplified commands
Add v2_main.rs and v2_commands.rs as thin wrappers over QpqClient from
quicproquo-sdk. Activated via --features v2 (feature-gated so v1 main.rs
remains the default). Includes auto-server-launch, clap-based command
surface (register-user, login, whoami, health, resolve, dm, group,
devices), and ClientConfig-based connection setup. MLS-dependent commands
(send, recv, group create/invite) print stubs pointing to the REPL.
2026-03-04 12:46:42 +01:00
4d62a837a5 feat(sdk): conversation management — DM/group lifecycle, outbox 2026-03-04 12:39:48 +01:00
67983c7a40 feat(sdk): OPAQUE auth module with state persistence 2026-03-04 12:39:33 +01:00
011ff541bb feat(sdk): messaging pipeline — send/receive with MLS, sealed sender, hybrid KEM
Full send pipeline: serialize_chat → MLS encrypt → sealed sender → per-recipient
hybrid wrap → batch/individual enqueue via v2 RPC.

Full receive pipeline: fetch/fetch_wait → sort by seq → hybrid unwrap → MLS decrypt
→ unseal → parse AppMessage. Includes retry loop for multi-epoch batches.
2026-03-04 12:39:15 +01:00
918da0c23d feat(sdk): key management, user resolution, device management
Add three new SDK modules wrapping v2 protobuf RPC calls:
- keys.rs: upload/fetch KeyPackages and hybrid public keys (5 methods)
- users.rs: resolve username <-> identity key (2 methods)
- devices.rs: register/list/revoke devices (3 methods)
2026-03-04 12:37:01 +01:00
6b757f8d65 fix(sdk): add missing session_token field to RpcClientConfig
The middleware-dev agent added session_token to RpcClientConfig
but the SDK client initializer was not updated, breaking the build.
2026-03-04 12:31:32 +01:00
d118fdbddf feat(server): v2 RPC handler dispatch for all 33 methods
Add v2_handlers module with ServerState, build_registry(), require_auth()
helper, and 33 protobuf handlers across 10 files:
- auth: 4 OPAQUE handlers (register start/finish, login start/finish)
- delivery: 6 handlers (enqueue, fetch, fetch_wait, peek, ack, batch)
- keys: 5 handlers (upload/fetch key package, upload/fetch hybrid key/keys)
- channel: create_channel
- user: resolve_user, resolve_identity
- blob: upload_blob, download_blob
- device: register, list, revoke
- p2p: publish_endpoint, resolve_endpoint, health
- federation: 6 stubs (Unimplemented)
- account: delete_account

All handlers decode protobuf, call domain services, encode response.
Auth handlers use full OPAQUE flow with session creation.
Delivery handlers include rate limiting and long-poll (fetch_wait).
2026-03-04 12:10:33 +01:00
6273ab668d feat(server): connection pool, session persistence, blob storage in SqlStore
- Replace Mutex<Connection> with Vec<Mutex<Connection>> pool (default 4)
  with try_lock fast-path and blocking fallback
- Add SessionRecord struct and session CRUD to Store trait (default no-ops)
- Implement session persistence in SqlStore (sessions table, migration 009)
- Add blob upload/download with SHA-256 verified staging assembly
  (blobs + blob_staging tables, migration 010)
- All 35 server tests pass, FileBackedStore unaffected
2026-03-04 12:09:03 +01:00
f09dbe10ce feat(rpc): auth handshake, server-push broker, audit logging
- auth_handshake.rs: connection-init protocol (magic 0x01, token, ack)
- push.rs: PushBroker manages per-identity push connections with gc
- server.rs: ConnectionState, auth handshake on first bi-stream, pass
  identity_key/session_token to RequestContext per stream
- client.rs: session_token in RpcClientConfig, auto auth handshake on connect
- middleware.rs: log_rpc_call with SHA-256 redaction, hex_prefix helper
- lib.rs: export auth_handshake and push modules
2026-03-04 12:08:20 +01:00
ff93275dc1 feat(server): complete domain service modules (keys, channels, users, blobs, devices, p2p, account) 2026-03-04 12:07:00 +01:00
a5864127d1 feat: v2 Phase 1 — foundation, proto schemas, RPC framework, SDK skeleton
New workspace structure with 9 crates. Adds:

- proto/qpq/v1/*.proto: 11 protobuf schemas covering all 33 RPC methods
- quicproquo-proto: dual codegen (capnp legacy + prost v2)
- quicproquo-rpc: QUIC RPC framework (framing, server, client, middleware)
- quicproquo-sdk: client SDK (QpqClient, events, conversation store)
- quicproquo-server/domain/: protocol-agnostic domain types and services
- justfile: build commands

Wire format: [method_id:u16][req_id:u32][len:u32][protobuf] per QUIC stream.
All 151 existing tests pass. Backward compatible with v1 capnp code.
2026-03-04 12:02:07 +01:00
394199b19b fix: security hardening — 40 findings from full codebase review
Full codebase review by 4 independent agents (security, architecture,
code quality, correctness) identified ~80 findings. This commit fixes 40
of them across all workspace crates.

Critical fixes:
- Federation service: validate origin against mTLS cert CN/SAN (C1)
- WS bridge: add DM channel auth, size limits, rate limiting (C2)
- hpke_seal: panic on error instead of silent empty ciphertext (C3)
- hpke_setup_sender_and_export: error on parse fail, no PQ downgrade (C7)

Security fixes:
- Zeroize: seed_bytes() returns Zeroizing<[u8;32]>, private_to_bytes()
  returns Zeroizing<Vec<u8>>, ClientAuth.access_token, SessionState.password,
  conversation hex_key all wrapped in Zeroizing
- Keystore: 0o600 file permissions on Unix
- MeshIdentity: 0o600 file permissions on Unix
- Timing floors: resolveIdentity + WS bridge resolve_user get 5ms floor
- Mobile: TLS verification gated behind insecure-dev feature flag
- Proto: from_bytes default limit tightened from 64 MiB to 8 MiB

Correctness fixes:
- fetch_wait: register waiter before fetch to close TOCTOU window
- MeshEnvelope: exclude hop_count from signature (forwarding no longer
  invalidates sender signature)
- BroadcastChannel: encrypt returns Result instead of panicking
- transcript: rename verify_transcript_chain → validate_transcript_structure
- group.rs: extract shared process_incoming() for receive_message variants
- auth_ops: remove spurious RegistrationRequest deserialization
- MeshStore.seen: bounded to 100K with FIFO eviction

Quality fixes:
- FFI error classification: typed downcast instead of string matching
- Plugin HookVTable: SAFETY documentation for unsafe Send+Sync
- clippy::unwrap_used: warn → deny workspace-wide
- Various .unwrap_or("") → proper error returns

Review report: docs/REVIEW-2026-03-04.md
152 tests passing (72 core + 35 server + 14 E2E + 1 doctest + 30 P2P)
2026-03-04 07:52:12 +01:00
4694a3098b docs: comprehensive update for sprints 1-9
Update README, ROADMAP, and mdBook to reflect all sprint deliverables:
rich messaging, file transfer, disappearing messages, Go/TypeScript SDKs,
C FFI, mesh networking (identity, store-and-forward, broadcast), and
security hardening. Add 6 new mdBook guides (REPL reference, Go SDK,
TypeScript SDK + browser demo, rich messaging, file transfer, mesh
networking). Check off 16 completed ROADMAP items across phases 3-9.
2026-03-04 02:10:20 +01:00