The subtree search returned early whenever it found *any* group member,
so an in-tree stage that was busy or sleeping on a timer masked a
reparented member genuinely parked on the tty.
Checking every member instead would have meant a /proc scan per session
per tick whenever nothing was parked — the common case. ProcGroups
snapshots /proc once per activity pass and indexes by process group, so
the cost is one walk however many sessions the daemon owns, and the
subtree walk disappears: the index is a superset of it.
Also fixes the background-helper test, which had stopped discriminating.
It polled until awaiting_input was true and caught a moment before the
helper spawned, so it passed against a deepest-descendant walk too. It now
waits for the helper to exist first. Both regression tests were re-checked
by patching each old implementation back in.
Reported in review on #51.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
tpgid names a process *group*, not a process. Its leader can exit while
another member still holds the terminal — a foreground pipeline whose
first stage finished — and then /proc/<tpgid> does not exist, so reading
it directly reported the session idle while it sat waiting on the user.
awaiting_input now searches the pane's subtree for live processes whose
pgrp matches, and asks whether any is parked on the tty. Members stay in
that subtree because their parent shell is alive, so they never reparent.
Reported in review on #51. The regression test needs an interactive shell
— job control is what puts a pipeline in its own process group — and waits
for the leader to actually be gone before asserting, otherwise it passes
against the old lookup while the shell is still foreground.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A plain deadline loop says the same thing without catching a panic to
turn it back into an assert.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
awaiting_input walked `children` to the deepest descendant, guessing that
the newest child is the foreground one. An agent that starts a background
helper and then blocks on read landed on the helper, so a session genuinely
waiting on the user was reported as idle.
tpgid in /proc/<pid>/stat is the kernel's own answer — the foreground
process group of the controlling terminal — and the group leader's pid is
the group id. One field read replaces a 64-deep walk.
Adds tests/activity.rs: the first end-to-end coverage of this pipeline,
which had none. The background-helper case fails against the old walk and
passes against tpgid.
Also makes the attach handler subscribe before upgrading. Checking
existence and then subscribing inside the callback left a window for the
liveness watcher to reap in between, which attach::serve compensated for
with a clean close; acquiring the subscription up front means the 404 and
the subscription are one step and the window does not exist.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
SPA attach had no reconnect at all. Any dropped frame closed the socket,
`onClosed` fired, and the pane was destroyed permanently — the TUI
reconnects forever on the same endpoint. Session death also reached the UI
on two uncoordinated channels: the row from the events socket, the pane
from `ws.onclose`, with nothing reconciling them, so either could happen
without the other.
The catalog is now the authority. `shouldRetryClose` decides from session
status: Running means the link dropped, so retry with 500ms-to-10s
backoff; Dead means the PTY stream was sealed and nothing more is coming;
absent means it is gone. Application closes (4000+, a relayed daemon
rejection) and clean closes are never retried. The shell also closes any
pane whose session has left the catalog. The policy lives in reconnect.ts
so it is tested without a DOM.
kill_session checked `has()` before `kill()`, which only opened a window
for the liveness watcher to reap in between — turning a 404 into a 502.
`kill` reports the miss itself.
with_spawn_history logged and carried on when `Arc::get_mut` failed,
producing a daemon whose recent-dirs picker is permanently empty with
nothing to show for it. It panics now: that is a construction-order
mistake, at startup, in code that runs once.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Each was a typed field that type-checked on both sides and carried
nothing across.
SessionStatus::Dead was never constructed by the daemon, yet the TUI, CLI
and SPA all render it and DESIGN.md specifies the exited badge. Sessions
were removed on death, never marked. They are now marked Dead and kept
until the operator dismisses them, with the PtyBridge sealed rather than
dropped: attached clients still get a Close frame (NAP-226), and the
replay buffer survives so a later attach can read what a crashed agent
printed. Previously that output was destroyed. Dead sessions are
read-only and do not survive a daemon restart.
CreateSessionRequest.env was accepted, stored, echoed back and dropped —
new_session_piped had no env parameter. It now reaches tmux as sorted
`-e KEY=VAL` flags. A key containing `=` or NUL is rejected, since tmux
would parse it as a different assignment than the caller wrote.
PROTOCOL_VERSION was written by the daemon and read by nobody. Adds
`maestro_protocol::is_compatible`, compared on the major component only;
`maestro list` shows the daemon version and the TUI warns on mismatch.
Also stops mapping every non-404 session error to 502: an invalid env key
is a client mistake and now answers 400.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A background task, a second broadcast channel and a whole module existed
to remap SessionEvent onto CatalogEvent one variant at a time.
SessionManager now emits CatalogEvent directly.
Deletes local_events.rs, the publisher task, the LocalEventBus channel,
enum SessionEvent, LocalApiState::events, and the duplicate capacity
constant — events_capacity was declared twice for one logical stream.
Lag semantics improve. There were two lag points: the publisher falling
behind the manager, which broadcast one Lagged to every client, and each
WS subscriber falling behind the bus. Only the per-subscriber one is left,
which pump_events already handled — a slow client now makes only itself
resync.
Adds the first daemon-level tests for WS /api/events: the Snapshot-then-
deltas contract, and that a late subscriber gets a Snapshot of current
state. Both were described only in a comment.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
SessionManager held a concrete Tmux, so its decisions — spawn rollback
ordering, liveness diffing, activity change-detection — could only be
reached by forking a real tmux server. Inner now holds an Arc<dyn Panes>;
Tmux is the production adapter, FakePanes the one tests drive.
Arc<dyn Panes> rather than a type parameter: a generic would spread
through LocalApiState, attach and cli, and one vtable hop is free next to
a fork+exec.
Adds tests/session_decisions.rs — six cases, no subprocess, 0.24s.
Replaces the chmod-0o500 rollback test, which detected root and returned
early, so on CI it asserted nothing; the failure is now forced by a
regular file sitting where the sidecar directory's parent must be, which
no uid can work around. It also asserts the rollback *order*, not just
the end state.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Rewrites CLAUDE.md as the current-architecture reference, drops the
registry from INSTALL/PRODUCT/DESIGN and the web README, deletes the
registry Dockerfile, and marks the dated design specs superseded where
they describe the registry, JWT auth, or cross-host attach.
The installer loses its configure step — every config field now has a
default, so a daemon starts with no config file at all.
Maestro is now single-host: one daemon owns the machine's sessions and
clients talk to it directly. Removes the registry crate, the daemon's
registry client/broker/catalog mirror, JWT attach auth, cookie auth, and
the perimeter (T2) transport config.
The daemon API is now unauthenticated on its bind address, same trust
model as tmux. Session and CreateSessionRequest lose daemon_id;
RegistryEvent becomes CatalogEvent with no daemon variants; attach moves
to WS /api/attach/:id with no token.
Replace the freeform `metadata.label` session grouping with nesting derived
from each session's working directory. The daemon's activity watcher already
polls each tmux pane every tick, so it now also captures
`#{pane_current_path}` and surfaces it as a new `Session.current_cwd`
(serde-default, back-compat); `cwd` stays the spawn dir for worktree
inference, spawn-similar, and recents.
Web catalog builds the tree from `current_cwd ?? cwd`, collapses the home
dir to `~`, and folds single-child folder chains so `~/git/maestro` reads as
one node. CLI swaps its LABEL column for an abbreviated DIR; the spawn forms
(web + TUI) and `--label` flag are removed.
The pane format stays space-delimited (parser splits the first three tokens
and keeps the remainder as the path) so a cwd containing spaces survives.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add an orthogonal activity dimension to sessions, inferred entirely from
structured tmux/`/proc` state — no PTY byte-stream parsing, no work on the
attach hot path.
Protocol: `ActivityState` enum (Active/Idle/AwaitingInput) + three additive,
serde-defaulted `Session` fields (`foreground`, `activity`, `last_activity_at`)
so the snapshot/diff flow stays forward-compatible with older daemons.
Detection (piggybacks the existing ~2s liveness tick, no new timer):
- foreground command via `tmux list-panes -F '#{pane_pid} #{pane_current_command}
#{pane_tty}'`;
- quiescence from a per-`PtyBridge` last-read `Instant`, bumped in `run_reader`;
- AwaitingInput via best-effort `/proc` walk to the deepest descendant: state
`S` + a non-timer `wchan` + an stdio fd on the pane tty. (`/proc/pid/syscall`
is ptrace-gated and unreadable for own sessions, so `wchan` distinguishes a
blocking read from a plain `sleep`.)
- combine: recent output → Active; else input-blocked → AwaitingInput; else
Idle. Emit `SessionUpdated` only on (activity, foreground) change — no diff
storms at steady state.
`maestro list` gains an ACTIVITY column and shows the live foreground. Pure
combine/parse logic is unit-tested; an e2e asserts `cat` → AwaitingInput and
`sleep` → Idle with correct foreground against real tmux.
Closes#18Closes#8
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Expose the daemon-side spawn data (#24 recents, #25 gitroot, #22 presets) to
the web, mirroring the NAP-329 ListDir broker plumbing end to end so it works
across federated hosts.
Wire (maestro-protocol): `RecentDir`/`RecentDirs`/`GitRoot` payloads;
`DaemonInbound::RecentDirs`/`GitRoot` requests + `ForwardRecentDirs`/
`ForwardGitRoot` relay pairs + their `*Result` replies on both enums. All
forward-compatible via the existing `RpcResult` + `#[serde(other)] Unknown`.
Relay (broker + registry + registry_client): home daemon → registry → owning
daemon → back, with the same fail-fast-while-disconnected + RAII pending-entry
cleanup discipline as ListDir. The owning daemon answers RecentDirs from its
spawn history and GitRoot by resolving the path against its home dir
(shared `dirlist::resolve`) then walking up via `gitroot::find_root`.
Daemon HTTP (maestro-daemon):
- GET /api/daemons/:id/recent-dirs → RecentDirs
- GET /api/daemons/:id/gitroot?path= → GitRoot
- Presets ride GET /api/web/info (the documented DaemonInfo sibling) so the
form needs no extra round-trip; resolved once at startup from [spawn].presets
(a bad preset list now refuses startup).
Tests: protocol round-trip + forward-compat for the new variants; broker
forward round-trips + fail-fast; HTTP handler tests (self recents/gitroot,
presets advertisement, peer 503 without a registry link).
Closes#26
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a `spawn_history` daemon module backing the spawn form's "recent dirs"
picker. Sidecars are deleted on kill, so they can't be the source — use an
append-only JSONL log at `<state_dir>/spawn-history.jsonl`, one record per
successful spawn (cwd, command, name, worktree_branch, at).
- Write: `SessionManager::create` appends after a successful spawn,
best-effort — a log-write failure only warns, never fails the spawn.
- Read: `recent_dirs()` folds newest-first, dedups by cwd, drops entries past
the recency cap, caps the result count.
- Prune: opportunistic `compact()` from `append` once the file passes a size
threshold; atomic tmp+rename rewrite.
- Config: `[daemon].spawn_history_days` (default 30, `0` disables). History is
wired into the manager at daemon startup.
Time-dependent logic is factored into `*_as_of(now)` helpers for
deterministic unit tests. Adds module unit tests (append/dedup/recency
prune/compaction/disabled/malformed/cap), a config resolution test, and
real-tmux integration tests proving create→append→read end to end.
Closes#24
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The peer (T2) attach path had a per-pair access-control layer
(`[daemon.policy].allow_inbound`): a target-side allow-list of daemon ids
permitted to broker-attach, enforced inside JWT verification, hot-reloadable via
SIGHUP / `POST /api/policy/reload`, with a `maestro policy` CLI. On a flat-trust
perimeter (the operator's VPN/firewall is the boundary for which hosts connect)
that granularity is unused, and deny-by-default made it config friction —
cross-daemon attach silently failed until every peer was listed.
Remove the allow-list; keep the registry-minted pair-attach JWT and ALL its
binding checks (iss=registry, aud=target, sid=session, exp). The firewall gates
host->port reachability; it does not stop a reachable host from attaching to a
session it wasn't handed or replaying a token elsewhere — that's the JWT's job
and it stays. Only the `sub in allow_inbound` check is dropped.
Removed: DaemonPolicySection / [daemon.policy].allow_inbound (config + daemon),
policy.rs (SIGHUP listener + GET/POST /api/policy), the `maestro policy` CLI,
PairAttachVerifier's allow-list field/reload/snapshot + JwtError::NotAllowed,
and the e2e broker_rejects_when_target_doesnt_allow_inbound test. The NAP-265
pre-upgrade-502 path stays covered by broker_attach_returns_502_on_perimeter_mismatch.
Also dropped the unit's ExecReload=kill -HUP: with no SIGHUP handler left,
default SIGHUP disposition is terminate, so wiring `systemctl reload` to it would
kill the daemon. Re-introducing a per-pair gate later is a verify-time change.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
CI runs as root, where the 0o500 read-only sidecar dir doesn't block the
write the test relies on failing — so it reported a spurious failure and
has reddened main since commit 149. Probe writability and skip when root;
non-root runs still assert the rollback.
Two web spawn QoL improvements plus the wire support they need:
- Command field now prefills `claude` instead of an empty box.
- Working directory prefills the selected daemon's home dir, updating on
daemon change unless the user has typed their own value.
The daemon resolves its `$HOME` via `dirs::home_dir()` at startup and
advertises it end-to-end: a new `home_dir` field on `DaemonOutbound::Hello`
flows through the registry's `register` into `DaemonInfo` (served on
`/api/daemons` and `/api/events`). The field is `#[serde(default)]`, so older
daemons that omit it deserialize to an empty string — backward compatible, and
the UI simply skips prefill when it's empty.
Hierarchical `/`-separated labels in the web sidebar already worked
(web/src/catalog.ts); no change there.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The home daemon's `POST /api/sessions` (peer `daemon_id`) and
`DELETE /api/sessions/:id` (peer-owned) now broker the op to the owning daemon
over the registry control link rather than returning 501/404. This fixes
cross-daemon spawn/kill for *all* local clients — notably maestro-web, which
talks only to the local daemon and had no registry fallback.
- broker: `BrokerHandle::{forward_create,forward_kill}` + completion handlers,
mirroring `resolve_attach` (fail-fast while disconnected, RAII pending
cleanup, RESOLVE_TIMEOUT).
- registry_client: route `Forward{Create,Kill}Result` back to the broker.
- local_api: create forwards on peer `daemon_id`; kill looks up the owner in
the catalog mirror and forwards (only a truly-unknown session is 404).
`broker_error_response` maps BrokerError → HTTP (503/504/502/404/...).
- cli: thread the broker into `LocalApiState`.
Tests: broker round-trip / fail-fast / registry-err propagation; updated the
local_api integration test (peer create now 503s without a registry link
instead of 501, proving it brokers).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
create() ran worktree-add → tmux spawn → sidecar write with no unwinding. A
failure in a later step leaked earlier ones — worst case a sidecar write
failure (disk full / EIO / permission) left a ghost `maestro-<uuid>` tmux
session running with no in-memory record, silently adopted with empty metadata
on the next daemon restart.
Each step now unwinds the ones before it: bridge/tmux failures remove the
worktree (+ FIFO), and a sidecar-write failure kills the just-spawned tmux
session before removing the FIFO + worktree.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
`SessionManager::recover()` re-inserted each sidecar's `Session` verbatim,
keeping the persisted `daemon_id`. If the operator renamed the daemon between
starts, recovered sessions were re-advertised under the old id and became
unroutable. Overwrite `daemon_id` with the daemon's current id on recover,
warning when the sidecar's id differs.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
`SessionManager::create` used to call `tmux new-session` (which exec's the
user's command immediately), run an async `PtyBridge::start_with_replay`
(mkfifo + open + spawn reader), then issue a *separate* `tmux pipe-pane`.
Any output the child wrote between exec and that third tmux subprocess
landing went into the pane but never into the FIFO. Hit in NAP-252
dogfood: `bash -c 'echo X; exec bash'` sessions showed only the trailing
prompt, never the X.
Two changes:
1. New `Tmux::new_session_piped(id, cwd, cmd, sink)` chains
`; pipe-pane -t maestro-<id> <sink>` into the same tmux command argv
as `new-session`. Within a tmux command chain, `pipe-pane` is
processed immediately after `new-session` returns — before the child
has finished `execve` — so the FIFO is wired up before the first
write.
2. `SessionManager::create` now builds the FIFO + reader task **before**
spawning the session, then calls `new_session_piped` instead of the
two-step `new_session` + `pipe_pane`.
`Tmux::pipe_pane` and the standalone `Tmux::new_session` stay around —
`recover()` (NAP-228) still needs a standalone pipe-pane re-attach, and
some integration tests want a bare session without a pipe.
Tests:
- Unit: argv assertion that the chained `pipe-pane` follows `new-session`
when a sink is supplied.
- Integration: `create_captures_initial_output_in_replay_buffer` spawns
`sh -c 'echo NAP_253_FIRST_LINE_MARKER; exec sleep 30'` and polls
`subscribe_pty_with_replay` until the marker shows up in the replay
buffer (or fails after 5s). Fails before the fix, passes after.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Capture sidecar_dir/fifo_dir/tmux/worktree_root before `drop(fx.manager)`
so the post-drop builder calls don't need to reborrow the partially-moved
fixture. Pure compile fix — same scenario as before.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`Tmux::pipe_pane` used `tmux pipe-pane -o`, which tmux documents as "only
open a new pipe if no previous pipe exists." When `maestrod` restarts and
`SessionManager::recover` re-issues `pipe-pane` against the adopted tmux
session, the previous daemon process's pipe to a now-orphan FIFO is
still active — so `-o` makes the call a no-op. The revived manager's
PtyBridge reads from a fresh FIFO that tmux never feeds, and attach
clients see input land in the pane but no output flow back.
Drop the `-o`. tmux closes any existing pipe before opening the new one,
which is what `pipe_pane` callers always wanted: `create` runs on a
fresh session (nothing to clobber), `recover` actively needs the
clobber.
Test: `recover_re_pipes_pane_so_output_reaches_subscribers` creates a
bash session under one manager, drops it, recovers under a second
manager, sends keys, and asserts the echo bytes reach a fresh
subscriber within 5s. Fails before the fix (stale pipe), passes after.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
CI e2e tests failed with "tmux exited with code 1: server exited
unexpectedly" because the startup `start-server \; set-option -g
default-terminal` call left an empty server that tmux's default
`exit-empty on` killed before the first `new-session` could attach.
Move the `set-option -g default-terminal tmux-256color` into the same
tmux command as `new-session` (chained with `;`) so a single tmux
invocation creates the server, sets the option, and spawns the pane.
The pane reads `default-terminal` at creation time, so chaining before
`new-session` still applies to the very first session.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`maestrod` running from a stripped systemd env leaves the spawned shell
without `TERM`/`COLORTERM`, and the pane process inherits tmux's
`default-terminal` (which defaults to `screen`, 8 colors). Browser
attaches via xterm.js render ANSI sequences as plain text.
Fix:
- `Tmux::new_session` now passes `-e TERM=xterm-256color -e
COLORTERM=truecolor` to `tmux new-session`.
- New `Tmux::set_option_global` chains `start-server \; set-option -g`
so it works before any session exists; the daemon calls it on startup
to set `default-terminal=tmux-256color`.
Operators wanting 24-bit color should add
`set-option -ga terminal-overrides ",xterm-256color:RGB"` to their
own tmux config — left out of the daemon since that's operator territory.
Tests:
- Unit: `new_session_args` returns the expected argv including the
`-e TERM=…` / `COLORTERM=…` pair.
- Integration: spawn a session under an isolated tmux server, set
`default-terminal`, run `printenv` + `tput colors`, assert the pane
sees `TERM=tmux-256color`, `COLORTERM=truecolor`, and 256 colors.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Each PtyBridge keeps a per-session ring buffer (Mutex<VecDeque<u8>>,
default 64 KB, override via `[daemon] pty_replay_bytes`). The reader
task appends each chunk while holding the buffer lock and broadcasts in
the same critical section, so `PtyBridge::subscribe_with_replay` —
which snapshots the buffer and subscribes to the broadcast under that
same lock — produces a snapshot that's a true prefix of subsequent live
bytes. No duplicates, no gaps at the seam.
`SessionManager` exposes the new `subscribe_pty_with_replay`, and
`attach::serve_local` sends the snapshot as a single binary frame
before pumping the live receiver. End result: attaching to an idle
bash session now shows the prompt immediately; reattaching after
running a build sees the last screenful instead of a blank terminal.
Adds `[daemon] pty_replay_bytes` to the unified config (None ⇒
default; Some(0) ⇒ disabled). E2E test exercises the full path: client
A drives output through, disconnects, client B sees the prior output
on connect.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
New module `local_events::LocalEventBus` (broadcast<RegistryEvent>) with
two publishers:
* `spawn_session_publisher` — translates SessionManager's own session
events into the matching `RegistryEvent` variants (with daemon_id
filled in for the SessionRemoved case).
* `CatalogMirror::apply` — re-publishes peer events after applying them
to the local mirror, so subscribers see the post-apply state.
New `WS /api/events` route on the daemon: subscribes to the bus, sends
an initial `Snapshot` built from manager.snapshot() ∪ catalog state,
then forwards live deltas. `Lagged` follows the registry's convention
(client resyncs via REST).
Revert the REST-seed hack in web/src/events.ts — the WS endpoint now
exists, so `CatalogStore::start()` can go back to just `connect()`.
E2E: subscribe to a daemon's events WS, expect Snapshot then a
SessionAdded after POSTing /api/sessions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* New PairAttachVerifier (crates/maestro-daemon/src/jwt.rs): checks
signature, aud == self, exp > now, and sub ∈ allow_inbound. Default
deny — empty allow_inbound rejects every cross-daemon attach.
* /sessions/:id/stream tries the legacy verifier first, falls back to
the pair verifier on mismatch. Compat with existing direct-attach.
* New BrokerHandle (crates/maestro-daemon/src/broker.rs): shared between
the registry-client task and HTTP handlers; calls ResolveAttach over
the control link and correlates the reply via a pending map.
* RegistryClient now owns a BrokerHandle, drains its outbound mpsc
alongside session events / heartbeats, and routes
ResolveAttachResult back into the pending map. First heartbeat fires
immediately so the registry has the daemon's perimeter info before
the first ResolveAttach.
* /api/attach/🆔 self-owned → local PTY; peer-owned (per catalog
mirror) → ResolveAttach + T2 dial + bidirectional frame proxy.
Returns structured errors for missing local perimeter, peer on a
different perimeter, etc. T1/T3 stay stubbed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
New module crates/maestro-daemon/src/catalog_mirror.rs: in-memory mirror of
the global catalog (peer daemons + their sessions), updated from the
`DaemonInbound::CatalogEvent` frames the registry now fans out.
`local_api` consumes it: `GET /api/sessions` returns the union of the
daemon's own sessions and the mirrored peers; `GET /api/daemons` returns
self + connected peers. The web UI sees the same catalog from any home
daemon.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
New module crates/maestro-daemon/src/local_api.rs serving:
* GET /api/sessions — own sessions
* GET /api/daemons — self only (v1; peer mirror is NAP-246)
* GET /api/web/info — daemon identity (browser header data)
* POST /api/sessions — create on self; 501 for cross-daemon (NAP-247)
* DELETE /api/sessions/:id — kill on self; 404 for unknown
* WS /api/attach/:id — delegate to existing PTY subscribe when self-owned,
404 otherwise (cross-daemon broker is NAP-247)
The new routes share bind_addr with the legacy /sessions/:id/stream WS
endpoint, which keeps working unchanged. The shared serve loop is exposed
as attach::serve_local so the broker WS can reuse the PTY pump.
Trust model documented at the top of the module: no JWT, no cookie on
the new routes — same trust as tmux being one-user-wide. Operators who
bind a perimeter address opt into one-segment-wide trust.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When a session terminates (the user runs `exit`, tmux is killed, or
`maestro kill` is called) the daemon's per-session broadcast channel
closes and `serve_attach` breaks out of its loop. The socket was then
dropped without a Close frame, so the client's `stream.next()` would
hang and the terminal stayed stuck in raw mode.
- daemon: send `Message::Close(None)` before returning from
`serve_attach` whenever the session itself ended (broadcast closed,
or session was killed between auth and subscribe).
- client: track remote-close, restore the TTY first, then print
`[maestro] session ended` so the message lands in cooked mode.
- tests: cover the kill path and the liveness-watcher reap path on the
daemon side.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The daemon never noticed when a session's command exited and tmux tore
the session down — the in-memory `SessionManager` map plus the registry
catalog kept showing it as `Running`, and a follow-up `maestro attach`
would hang on an empty broadcast.
Adds a per-host background task that polls `tmux list-sessions` every
2s, diffs against the in-memory set, and reaps anything we still track
but tmux has lost: drops the record, deletes the sidecar + FIFO,
cleans up its worktree if requested, and emits `SessionEvent::Removed`
so the registry purges its catalog.
The task holds a `Weak<Inner>` so it exits cleanly once the last
`SessionManager` clone is dropped. Skips the tmux subprocess when the
in-memory map is empty.
Adds the client-facing surface on the registry (NAP-215). Endpoints:
- GET /api/daemons — list connected daemons (with daemon_url)
- GET /api/sessions — flat session catalog
- POST /api/sessions — forwards CreateSession to chosen daemon, 504 on
10s daemon timeout
- DELETE /api/sessions/:id — forwards KillSession
- GET /api/sessions/:id/connect-info — returns {daemon_url, jwt} with a
5min HS256-signed attach JWT
- WS /api/events — snapshot + delta broadcast of catalog and presence
- GET / placeholder served until NAP-218 web bundle lands
Cookie auth middleware mints a long-lived httpOnly JWT cookie on first
request and reissues silently on missing/tampered cookies. Per spec the
perimeter is the operator's network — the cookie is a second factor.
Protocol additions: daemon_url on Hello, DaemonInfo, ConnectInfo,
RegistryEvent. Daemon config + RegistryClient now carry daemon_url.
Registry state now exposes an event broadcast channel (subscribe gives
a snapshot + tokio::sync::broadcast::Receiver) and helper lookups
daemon_for_session / daemon_url.
Tests: 11 reqwest-driven HTTP integration tests cover every endpoint,
the cookie flow, 504 timeout path, worktree forwarding, and events WS
propagation.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Completes the headless daemon stack for NAP-214.
- `config`: daemon TOML config (daemon_id, registry_url, bind_addr,
token, state_dir, worktree_root) with sensible XDG defaults.
- `jwt`: HS256 verifier; the daemon validates per-attach JWTs locally
with no callback to the registry. Also exposes `issue()` for tests.
- `pty::PtyBridge`: per-session FIFO + tokio broadcast pump. The FIFO is
opened r+w so it never sees zero writers — stays alive across
`pipe-pane` re-issues. Helper `pipe_pane_command()` builds the
`cat >> <fifo>` shell line.
- `session::SessionManager`: authoritative live-session store. Creates
worktrees (when requested), spawns tmux, mkfifo + pipe-pane, writes
sidecar; tears it all down on kill. Emits diff events on a
broadcast for the registry client. `recover()` adopts surviving
`maestro-*` tmux sessions on startup and re-pipes them.
- `registry_client::RegistryClient`: persistent outbound WS to the
registry with exponential backoff. Sends Hello → snapshot, drives a
select loop that pumps inbound CreateSession/KillSession RPCs into
the manager (replying with results), emits SessionAdded/Removed/
Updated diffs from the manager, and heartbeats every 10s.
- `attach`: axum router mounting `WS /sessions/:id/stream?token=<jwt>`.
Binary frames carry PTY bytes either way (subscribe to broadcast →
send to client; receive from client → tmux send-keys -l). Text
frames carry `ClientControl::{Resize,Ping,Pong}`.
- `bin/maestrod`: ties everything together, loads config from
`MAESTROD_CONFIG` env or `~/.config/maestrod/config.toml`,
recovers, spawns the registry client, binds the WS server.
Also adds `JwtClaims` to `maestro-protocol` so the daemon and the
future registry JWT issuer (NAP-215) agree on the wire shape.
Integration tests cover session manager round-trip + recovery + PTY
broadcast; attach endpoint JWT (bad/wrong-session/wrong-daemon),
PTY round-trip through real tmux, multi-attach broadcast, ping/pong;
registry client registration + snapshot + diff forwarding +
CreateSession RPC round-trip.
Closes NAP-214.
Local primitives for `maestrod`, no network code yet:
- `tmux`: subprocess wrapper around new-session, kill-session,
list-sessions (filtered to maestro-*), send-keys -l, pipe-pane -o,
resize-window, kill-server.
- `sidecar`: atomic JSON metadata store keyed by session UUID, with
load_all for daemon restart recovery.
- `worktree`: `git worktree add -b` with collision-proof generated paths
and `git worktree remove --force` for cleanup_on_kill.
Integration tests use real tmux (isolated via unique socket + temp
TMUX_TMPDIR) and real git (tmpdir repo).
Closes NAP-212.
