Commit Graph
105 Commits
Author SHA1 Message Date
Miguel Palhas ae3091e2a2 fix(maestro): guard resume against kill and second resume
The respawn runs with the session map unlocked, so two takes of the
lock straddle it. A kill landing in that window deleted the sidecar
before resume wrote it back, leaving a file for a session that was
gone — the next recovery registered it as a dormant ghost. Resume now
deletes the sidecar in the branch where it finds its own record gone.

Two resumes both passed the `Running` check for the same reason. The
record is marked in-flight under the first lock, so the second request
is refused with `already_running` instead of losing a race for the
FIFO and reporting a tmux failure. Every path out of resume clears the
marker.

Both races are driven deterministically in the tests: the manager's
tmux spawn is parked mid-resume while the racing kill or resume runs.

Closes #175
2026-09-04 09:19:59 +01:00
Miguel Palhas d050df7d3d fix(maestro): always probe tmux during recovery
#164 skipped the tmux probe whenever the stored boot id differed from the
current one, assuming a changed boot id proves the tmux server is gone. It
does not: a restored state directory, a VM or container snapshot rollback,
or a truncated boot_id file all change it while tmux keeps running with
live sessions. Every one of those sessions was then registered Dormant and
stranded — resume respawns `maestro-<id>`, which collides with the name
tmux still holds, so it failed 502 forever.

Recovery now always runs `tmux list-sessions`, the single source of truth
for liveness the rest of the daemon assumes. The boot id keeps its other
job: telling a reboot from a restart in recovery's log.

The #164 test asserted the old behaviour with tmux still alive, so it is
replaced by one asserting the session is adopted Running, plus one covering
a real reboot — boot id changed and the tmux server dead — still going
dormant.

Refs #174
2026-09-04 09:13:15 +01:00
Miguel PalhasandClaude Opus 5 02bf077c03 feat(maestro): add [daemon].resume_on_boot policy
Recovery now has a stated policy for sessions whose tmux pane is gone but
whose sidecar survived: dormant (default), auto, or never. auto goes down
the same SessionManager::resume the client's POST takes, so there is one
resume path. never prunes the sidecars the way #163 does for a missing cwd.

The default is dormant because resuming agents unattended seconds after a
reboot is a foot-gun; auto is opted into per host. Recovery logs which
policy ran and how many sessions it acted on.

Refs #167

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-04 09:06:51 +01:00
Miguel Palhas a793727bb9 merge: resume a dormant session in place (#166)
Closes #166
2026-09-04 08:59:37 +01:00
Miguel Palhas bb6b885c68 feat(maestro): add POST /api/sessions/:id/resume
A dormant session comes back under the same uuid and sidecar, so its
name, created_at, env and worktree ownership survive; only the tmux
session and the PTY bridge are new. The command is the matching preset's
resume_command with {agent_session} filled in, falling back to the
command the session was spawned with.

Refused with 409 for a session that is already running or whose cwd is
gone. Attaching to a session with no bridge now answers 409
not_attachable instead of 404 unknown_session, so a client can tell a
resumable session from one that does not exist.
2026-09-04 08:53:06 +01:00
Miguel PalhasandClaude Sonnet 5 b05b014e84 fix(maestro): clear worktree from the session clone on natural exit
mark_session_dead cleared blocked_work on the local session clone
written to the sidecar, but not worktree — so a pristine tree the
daemon just removed stayed named in the sidecar, and recovery from
that sidecar advertised a branch and path that no longer exist.

Refs #172.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-09-04 08:48:24 +01:00
Miguel Palhas ca2b60c17e merge: detect host reboot via boot_id (#164)
Closes #164
2026-09-04 08:43:11 +01:00
Miguel Palhas d548d0b24e feat(maestro): skip the tmux probe on a detected host reboot
Recovery could not tell a daemon restart from a host reboot: both
looked like "some sessions were not in tmux list-sessions", so the
tmux probe ran unconditionally and the logs said nothing useful when
it was actually a reboot.

Record /proc/sys/kernel/random/boot_id in the state directory on
every recover(). When it differs from the stored value, every session
from the previous boot is provably dead, so the tmux probe is skipped
and the second recovery pass registers them dormant directly, logging
"host rebooted, sessions dormant" with the count. A missing stored
value or an unreadable boot id (non-Linux, restricted /proc) falls
back to today's probe-based recovery.

The boot id source is injected via a BootIdSource trait so tests can
simulate a changed or unreadable boot id without rebooting the
machine.

Refs #164
2026-09-04 08:40:10 +01:00
Miguel Palhas aa34d975a9 test(maestro): assert the agent session id survives recover()
The test named for recovery only read the sidecar back; nothing covered
the id reaching the recovered session, which is what a resume reads.

Also narrows RESTART.md's leniency claim: the custom Deserialize lives in
the client's own binary, so an older build still fails on Dormant.
2026-09-04 08:36:43 +01:00
Miguel Palhas c004b5d1b6 feat(maestro): prune sidecars whose cwd no longer exists
recover()'s dormant-registration pass (#162) now deletes a sidecar
instead of registering it dormant when the directory it points at is
gone. Nothing can be resumed there, so the honest move is to remove
the record rather than keep an unactionable one around (#163).

Sidecars whose cwd still exists are unaffected, regardless of age.
2026-09-04 08:30:39 +01:00
Miguel Palhas c51d5b8f7b merge: persist agent session id (#161)
Closes #161
2026-09-04 08:27:49 +01:00
Miguel PalhasandClaude Opus 5 72644dd564 feat(maestro): register dormant sessions on recovery
A host reboot kills the tmux server, so `SessionManager::recover` iterated an
empty list and every session vanished from the API while its sidecar sat
untouched on disk. A second recovery pass now reads the sidecars tmux had
nothing to adopt and registers each as `SessionStatus::Dormant` — a record of
what was running, with no pane and no PTY bridge behind it.

`SessionStatus` deserializes leniently now, like `ActivityState`: an unknown
spelling reads as `Dead` rather than failing the whole `Session`.

Refs #162

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-04 08:22:52 +01:00
Miguel Palhas fa846000e7 feat(maestro): persist the agent's own session id on the sidecar (#161)
The Claude Code hook already receives a session_id in every payload, but
the activity route dropped it and the daemon had no field to store it on.
Parse it out of the hook body, add Session.agent_session_id, and write it
through to the sidecar once — the first report that carries one wins, and
later reports never overwrite it since an agent's own id never changes
mid-session.

This is the input to a follow-up POST /api/sessions/:id/resume.
2026-09-04 08:22:51 +01:00
Miguel Palhas 8325bc6d9a Merge remote-tracking branch 'origin/main' into blitz/command-bar
ci / web (push) Successful in 37s
ci / flake (push) Successful in 2m52s
ci / rust (push) Successful in 5m44s
2026-09-03 22:58:41 +01:00
Miguel PalhasandClaude Opus 5 8f499dbee3 fix(maestro): arbitrate pane size across clients
ci / web (push) Successful in 52s
ci / flake (push) Successful in 2m35s
ci / rust (push) Successful in 5m57s
Every attached client pushed its own dimensions straight to
`tmux resize-window`, so two clients of different sizes resized the
shared window past each other on every report — 198 resizes in 90
seconds on a four-session daemon. Each flip repaints a full-screen agent
at a new height and orphans the rows it drew at the old one, leaving a
stale input box the operator types into and reads as a stuck pane.

The window now goes to the smallest attached client per axis, tmux's own
rule for multiple clients, and only when that size actually changes.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-03 22:28:40 +01:00
Miguel Palhas ba02b11f13 feat(maestro): serve the configured leader to the web client
The TUI read its leader from `[tui].leader`; the web used `DEFAULT_LEADER`,
a module constant. Rebinding one left the other on `ctrl-a`, so the two
escape hatches hanging off the leader (`leader ctrl-k`, `leader leader`)
moved on one client and not the other.

`GET /api/info` now carries `leader`. The web binds that when it arrives and
rebuilds the leader-owning `CommandSet`, so the `?` overlay and the command
bar render the configured chord; an absent or empty field — an older daemon,
or an unreachable `/api/info` — leaves it on `ctrl-a`.

Chose the existing `[tui].leader` served as-is over a new shared key: one
knob, no precedence rules between two spellings, and no migration for
operators who already set it. The name is a wart — it is the leader for every
client now, not the TUI's — and `docs/COMMAND-BAR.md` and `docs/INSTALL.md`
say so.
2026-09-03 18:33:01 +01:00
Miguel Palhas 9d9aef119d feat(maestro): stamp last-focused time on the session (#145)
Add Session::last_focused_at, shipped on the wire like every other
field. Both POST /api/sessions/:id/focus (focused: true) and opening
/api/attach/:id stamp it and emit SessionUpdated, so a client that
only attaches sorts the same as one that reports focus explicitly.
Losing focus does not clear it, and a never-focused session has None,
which is what lets the command bar's empty-query ordering
(docs/COMMAND-BAR.md) sort it after every session that has focused.
2026-09-03 14:05:36 +01:00
Miguel Palhas 9dad3e6472 Merge remote-tracking branch 'origin/main' into animato
ci / web (push) Successful in 22s
ci / flake (push) Successful in 4m47s
ci / rust (push) Successful in 5m59s
# Conflicts:
#	crates/maestro-daemon/Cargo.toml
#	crates/maestro-daemon/src/local_api.rs
#	crates/maestro/Cargo.toml
2026-09-02 18:37:53 +01:00
Miguel PalhasandClaude Opus 5 4fc90aabfb feat(maestro): serve the web UI from the daemon
The browser reached the daemon through maestro-web, a second binary that
served the SPA and reverse-proxied /api/* — two TCP connections and a
WebSocket frame re-encode on every keystroke and every byte of PTY output.
On one machine that hop buys nothing.

The daemon now embeds the bundle and serves it off its own listener, behind
a default-on `web` feature. `--no-default-features` builds without pnpm or
node; MAESTRO_WEB_DIST still takes a prebuilt bundle.

The router's fallback answers an unmatched /api/* with 404 rather than the
SPA shell, so a client expecting JSON gets a status it can read instead of
HTML.

Removes maestro-web and the [web] config section. An existing [web] block
is ignored, not rejected. The installer deletes a leftover
maestro-web.service on upgrade.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-02 18:23:18 +01:00
Miguel Palhas e7b323b412 feat(maestro): name a session from its first prompt
POST /api/sessions/:id/prompt claims a session's one naming attempt and runs
claude -p --model haiku in the background; the answer is reduced to a short
slug and replaces the generated handle, settling the session at Haiku. A
chosen name, a rename mid-run, a missing binary or a timeout all leave the
handle alone. The prompt goes to the child on stdin and is never logged.

Refs #131
2026-09-02 17:57:17 +01:00
Miguel Palhas 3dee4a4e47 feat(maestro): daemon assigns the session handle
The spawn form no longer prefills a name, so create() fills one in: a
requested name is marked User, an absent one gets a generated handle marked
Generated. Rename always settles to User.

Refs #131
2026-09-02 17:49:09 +01:00
Miguel PalhasandClaude Opus 5 eaf60d9dac feat(maestro): stamp the daemon with its build commit
`GET /api/info` now carries `build` (commit hash, commit date, dirty flag)
and `instance`, a fresh uuid per daemon process. Together they let a client
name the release it is talking to and, after losing its socket, tell an
upgrade from a plain restart from a network blip.

The stamp comes from `maestro-daemon/build.rs`, which reads the git checkout
it compiles in. The Nix build has no `.git` in the crane-filtered source, so
`flake.nix` passes the flake's own revision and last-modified date through
`MAESTRO_BUILD_*` instead. Neither available leaves the fields empty.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-02 15:40:41 +01:00
Miguel PalhasandClaude Opus 5 c9cfdb5668 feat(maestro): one-word handles, quieter project boxes
Session handles are one word from a single 218-entry list — instruments,
forms, notation and theory, tempo and dynamics — replacing the
adjective-noun pair. A batch is shuffled, so it repeats only once the
wordlist runs out.

Project boxes show the parent directory instead of the full path: the
last segment is already the name line above it. The last command is gone
from the box, and picking a project no longer pre-selects it — the
directory and what runs in it are independent choices. Both changes land
in the TUI picker too, which mirrors the web form's rules.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-31 22:40:49 +01:00
Miguel Palhas eb9b4b4ae8 fix(maestro): replay pane mouse modes on attach
A full-screen app enables mouse reporting once, at startup. A client
attaching later never saw those sequences, so the web terminal read the
pane as a plain alternate screen with mouseTrackingMode "none" and its
touch and wheel gestures had nowhere to go — scrolling did nothing in a
browser opened after the session started.

Read the pane's mouse flags in the same display-message call and emit
the matching DECSETs ahead of the screen content. tmux expands an
unknown format name to the empty string, so the parser only trusts the
mouse tail when every field is present.

Closes #128
2026-08-31 22:07:22 +01:00
Miguel PalhasandClaude Opus 5 c98c11b5c3 fix(maestro): replay tmux's screen to fresh attaches
A new attach replayed the tail of the raw byte buffer. That tail starts
mid-stream, long after a full-screen program set the modes, colours and
cursor position that give those bytes meaning, so a browser refresh
painted garbage over every agent TUI until something forced a repaint —
which is why opening the same session in the TUI (a resize) fixed it.

Ask tmux for the pane's current screen instead, and fall back to the byte
buffer when there is no pane left to capture, as for a dead session.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-31 09:37:13 +01:00
Miguel PalhasandClaude Opus 5 d39fbdbbfc feat(maestro): drop retained worktrees and header debug toggle
Removes the retained-worktree store end to end: the daemon's state-dir
store, GET/DELETE /api/worktrees, `maestro worktrees list|reclaim`, the
TUI overlay behind `w`, and the web `#/worktrees` view and header link.

The kill guard stays. A forced kill and a dirty natural exit now leave
the worktree on disk untouched instead of parking it in a store, so
removing one is an ordinary `git worktree remove`. A pristine tree is
still removed on exit. DELETE /api/sessions/:id always answers 204.

The header's debug button also goes; Ctrl+Shift+D and `?hud` still
toggle the HUD.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-31 09:00:52 +01:00
Miguel PalhasandClaude Opus 5 345650172b feat(maestro): rename a session over PATCH /api/sessions/:id
The daemon trims and caps the name, persists it to the sidecar under the
sessions lock, and broadcasts SessionUpdated. A blank name clears
metadata.name so the card falls back to the command, matching the rule the
spawn form already uses.

Refs #125

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-30 11:26:12 +01:00
Miguel Palhas bbafff5a98 fix(maestro): add settle-finished observable for natural-exit test
The no-retained-store test asserts a negative — the tree survives — so
waiting on that negative proves nothing: settle could still remove the
tree one instant after the assertion passes. mark_session_dead now flips
a per-record settled flag once the exit rule has been applied or refused,
exposed as SessionManager::is_settled; the test waits on it before
asserting.

Refs #115
2026-08-24 23:12:46 +01:00
Miguel Palhas f689c916bf fix(maestro): poll awaiting-input until leaderless state holds
The pipeline leader exits as soon as both stages are forked, which can
beat the surviving stage opening the tty and parking on read — the one-
shot assertion then ran against a group with no parked member yet.
Poll for the answer, accepting it only while the foreground group is
still leaderless so a shell that regained the tty cannot satisfy it.

Fixes #113
2026-08-24 22:50:02 +01:00
Miguel Palhas 88233cb96d fix(maestro): sync watcher re-emit test to real event flow
Two races. First, the watcher updates the snapshot and emits
SessionUpdated as two steps, so the drain could run between them: the
legitimate breakdown emission then arrived inside the quiet window.
Subscribe before the watcher starts and consume through the breakdown
event so the window begins after the last real change.

Second, the assertion treated any SessionUpdated as a failed
suppression, but the activity pass may legitimately announce an
Active→Idle transition in that window (ACTIVE_WINDOW is 3s), carrying
today's blocked_work unchanged. Only a changed indicator counts as a
re-emit now.

Fixes #113
2026-08-24 22:50:02 +01:00
Miguel Palhas efbccb4e11 fix(maestro): wait for settle in pristine natural-exit test
Dead status is observable a moment before the settle step removes the
worktree, so the assertions raced the removal under parallel load. Wait
for the tree to be gone and the store to stay empty, same as the
retained-store wait cfebfd4 added to the guarded sibling.

Fixes #113
2026-08-24 22:41:54 +01:00
Miguel Palhas ea3e18d1f2 merge: sidebar-cards/111-flaky-retention 2026-08-24 22:11:30 +01:00
Miguel Palhas cfebfd441b fix(maestro): wait for retained-store write in natural-exit test
Dead status can be observed before the settle step's store write
lands, so the test's store read raced the write under parallel load.
Wait for the store to have the entry before asserting on it, same as
the settle-state wait a few lines below.

Fixes #111
2026-08-24 22:03:56 +01:00
Miguel Palhas 5a594b2b7f feat(maestro): serve POST /api/sessions/:id/activity 2026-08-24 21:38:42 +01:00
Miguel Palhas 8370ddf066 feat(maestro): hold agent-reported activity above inference for a TTL 2026-08-24 21:36:56 +01:00
Miguel Palhas 5c4e083c2d feat(maestro): add activity_detail to Session and the activity-report wire types 2026-08-24 21:32:27 +01:00
Miguel Palhas 002c128ac8 feat(maestro): stamp activity_since only on state transitions (#104)
The activity pass now sets activity_since only when the decided
ActivityState differs from the stored one; foreground/cwd-only diffs and
steady-state ticks leave the clock alone, so the sidebar duration counts
time in the current state instead of reading zero.

Daemon restarts still reset the clock to the restart time (stated in
recover()); the daemon cannot know how long a recovered session had
already been idle.
2026-08-24 19:49:55 +01:00
Miguel Palhas 90b7953b17 merge: sidebar-cards/103-session-env 2026-08-24 19:31:35 +01:00
Miguel Palhas b3c73fd2cb test(maestro): assert the exported session env end to end through real tmux 2026-08-24 19:30:11 +01:00
Miguel Palhas 7d06bec21e feat(maestro): export MAESTRO_SESSION_ID and MAESTRO_URL into spawned sessions
Every tmux-spawned session now gets its UUID under MAESTRO_SESSION_ID and,
when the runner has wired its base URL from [daemon].bind_addr, that URL
under MAESTRO_URL (#103). The daemon-owned values override anything the
spawn request carried. Adopted sessions stay without them by design.
2026-08-24 19:30:11 +01:00
Miguel Palhas b95b5c04c1 feat(maestro): add activity_since to the session wire 2026-08-24 19:19:33 +01:00
Miguel Palhas 2dcef860a6 fix(maestro): group worktree sessions under their repository
The sidebar nested a worktree session under the worktree, so a repo grew
a <repo>/worktrees/<branch> folder holding one session of the same name,
and the repo stopped being where you look for the work.

The daemon recorded whatever base_repo the request named, which is the
worktree itself when spawning from inside one; it now records the
resolved main checkout. The web tree groups by that instead of the cwd.
The TUI sidebar is a flat list with no directory grouping, so it needs
no mirror.

Closes #99
2026-08-24 17:17:35 +01:00
Miguel Palhas 187bcb01d6 feat(maestro): refresh the blocked-work indicator in the watcher (#83)
run_blocked_work_pass rides the liveness tick but probes on a slower
cadence (DEFAULT_BLOCKED_WORK_INTERVAL, 15s): each probe is three git
subprocesses per session. Sessions without a worktree are filtered out
before any probe, so their indicator costs nothing. A failed probe keeps
the previous value; settled exits clear it.
2026-08-24 13:04:30 +01:00
Miguel Palhas 003e3839bc feat(maestro): carry a blocked-work indicator on Session (#83)
Session.blocked_work is Option<InTheWay>, refreshed by the daemon's
watcher; None for worktree-less sessions and until first probed.
Defaults on the wire so older payloads stay forward-compatible.
2026-08-24 12:58:16 +01:00
Miguel Palhas 8aa007c73b fix(maestro): drop the cleanup_on_kill call sites #87 could not reach
Three in session_manager.rs arrived with #82 after #87 branched; four in
maestro-e2e were fenced off because #88 owns that crate. Removing the
field without them left the workspace uncompilable.
2026-08-24 12:12:02 +01:00
Miguel Palhas f4d12fa02d merge: drop cleanup_on_kill from the wire and the CLI
Closes #87
2026-08-24 12:11:31 +01:00
Miguel Palhas 31702e03e3 refactor(maestro): drop cleanup_on_kill from the wire and the CLI
Nothing has honoured WorktreeSpec.cleanup_on_kill since #56, and after
#79 and #81 survival is decided by what is in a worktree, not by what
the spawn asked for. A field that lies is worse than a missing one.

Removes the field from the protocol, the --cleanup-on-kill flag from
the CLI, and the tests asserting its default.
2026-08-24 12:08:41 +01:00
Miguel Palhas 2fc19a886e feat(maestro): apply the kill's worktree rule when a session exits on its own
When the liveness watcher finds the tmux session gone, it now settles
the worktree through the same probe_worktree/record_retention helpers
the kill uses: pristine trees are removed, anything else is retained as
exited-with-unsaved-work before anything is torn down. The dead session
drops its worktree reference once settled, so dismissing it neither
refuses nor retains twice. If the probe or store write fails, nothing
is destroyed and the tree stays attached to the record.

WORKTREES.md: written down the settle-then-drop and nothing-destroyed-
when-unrecordable rules alongside the existing exit rule.
2026-08-24 11:54:43 +01:00
Miguel Palhas b96f18c7cb feat(maestro): forced kill ends the session and retains the worktree
--force overrides the kill refusal and nothing else (#81). The guarded
worktree is recorded in the retained store with reason force-killed —
never removed — and the outcome names what was kept so a client can say
where the work went. A tree whose guard is clear never needed the
override; force removes it as usual.

The store write happens before teardown, so a failed retention leaves
the session running and every artifact intact. DELETE /api/sessions/:id
takes ?force=true and answers 200 with KillOutcome; the unforced kill
keeps its 204.
2026-08-24 11:29:45 +01:00
Miguel Palhas 62581d9f85 merge: refuse a kill when the worktree holds unsaved work
Collapses the two probes #79 and #80 grew in parallel into one:
worktree_probe::in_the_way is the single predicate behind the kill
refusal, the retained list, and the natural-exit classifier to come.
Keeps #79's in-worktree resolution (detached HEAD, missing directory)
and #80's InTheWay payload, and drops the duplicate UnsavedWork type.

Closes #79
2026-08-24 11:09:24 +01:00