Dev Pipeline panel (part 2)
The second half of the Dev Pipeline panel page: the Auto-lander view inside Runs, the QA Fleet view inside Stages, and the Setup tab — the switches, the cleanup jobs and the spend caps.
On this page
What it is
This is part 2 of the Dev Pipeline panel page. That page covers what the panel is, where it lives, and its rail — the three entry points Runs, Stages and Setup, including the Worktrees half of the worktree view. This half carries the views Runs and Stages open — the Auto-lander (from Runs) and QA Fleet (from Stages) — and the Setup tab.
Where to find it
Both halves are the same surface: the built-in Dev Pipeline virtual project, which docks in the main panel beside the projects sidebar and opens from its own row in the Omniscio sidebar. On a phone the rail becomes a row of tabs across the top, so the views described here are reached the same way remotely.
How it behaves
Auto-lander. A sub-view of Runs (reached from a quiet line there, not a rail tab of its own), this is where you watch it: live daemon status, the branch-life list — each ready branch with where it is, what it waits on, how long, and whether you're needed — and the land history — every branch it has landed, handed back on a conflict, or set aside (couldn't land), kept across app restarts, plus two quick actions: pause/resume the whole lander, and pause/resume one repository. Full write-up: auto-lander-dashboard.md. Its switches live in Setup, under All settings: turn it on or off, and choose which repositories it watches (setting a repo to observe-only first). The auto-lander lands branches you've marked ready into your main branch automatically and locally; it only ever moves things forward and never pushes. A branch you mark ready is trusted as finished — the lander merges it even if the dev-pipeline's process record was never marked complete (that check is informational, so a finished branch never sits stuck waiting for a record that was never updated). If a branch you marked ready gets more commits after tagging — or is rebased onto a newer main — the auto-lander now re-checks and re-tags it for you so it still lands, instead of stranding it behind a "moved after tagging" alert; you're only notified if that automatic re-check genuinely can't land it. A branch that turns out to have nothing left to land — it already landed, or its work reached main another way — is not re-tagged: the same automatic pass retires it through the app's delete door, which saves the branch's tip to a backup ref first (copied into a verified archive bundle before that ref is ever cleared), and it finishes without a card. When a branch lands you also see it in two more places: the session that produced it gets a green "Auto-landed …" note (with a check) just below its final message, naming the branch and your computer's main branch — even after that session has finished (it's never reopened or re-run) — and the run is marked "Landed ✓" in the Stages tab instead of just quietly disappearing when its worktree is cleaned up. If the auto-lander's integration checkout ever drifts out of sync it quietly re-syncs it — this self-heal runs silently (logged for diagnostics, never an inbox notice) and always snapshots a backup first, so nothing is lost.
Expanding a conflict row lists the conflicting files. That list is bounded and says so: the first 20 with a "+N more" tail, and — because only the first 50 conflicting paths are ever recorded per conflict — a plain note when you're at that recording limit, so a 90-file conflict never masquerades as a complete 50-file one.
Automatic master catch-up. The other thing in this panel that can move your local master
on its own, which is exactly why it sits beside the auto-lander: one background job — in the
app, every 15 minutes, and only when you run Omniscio from source — that fetches the remote
and merges it into your local master for you. Off by default. Off does not mean silent:
you are still told when you have fallen behind, with a one-click catch-up — nothing moves
your branch unless you ask. Turning it on is an ordinary setting, so an agent can turn it on
for you over the control server like any other setting.
What a tick actually lands. Each tick works for free first — merging what the deterministic
ladder can prove (fast-forwards, clean merges). A file that genuinely needs judgment is not
waited on: the merge still lands on your local master at once, with each such file held at the
trunk's copy rather than the merge's attempt at it — and which copy that is depends on the
computer. The trunk is GitHub on every computer except the main computer (the one computer
set to publish to GitHub — the Setup tab's main-computer switch). So on an ordinary computer a held
file shows GitHub's version and your own changes to it are banked and owed; on the main
computer your own master is the trunk, so a held file keeps your version and GitHub's
side is the part held back. Either way, paid AI
sessions are then bought, one per group of related files, to settle the held files — each group
lands the moment its files pass the lost-work check. So git log can show the remote's commits on
your master while a held file still lacks one side: GitHub's on the main computer, your own on
every other.
That paid work is bounded by a daily cap on those merge sessions (100 by default; 0 stops the group buyers outright, so held files simply stay held until you raise the cap). The cap counts those group buyers only — starting a session to finish a sync that was held from an earlier run, or to replace a worker that stopped, is not counted against it, so 0 on its own does not mean nothing can spend. Your chosen number is read from your settings each tick, and a read that fails falls back to 100, never 0. That is one switch, not two: it replaces the old pair (a free catch-up timer and a separate paid hourly sibling), because the free path alone kept stalling on the conflicts it could not settle. It ships off, and you turn it on here in the panel.
QA Fleet. The QA fleet's results, in the panel — what the background QA runs found, read
from the ledger they write locally (~/.amc/qa-fleet/ledger.db). It is a sub-view of Stages,
opened from a quiet line there that reads "QA fleet — N open findings" whenever there is at
least one, so you can see at a glance whether there is anything worth opening.
Five totals lead the view — Open Findings, High/Critical Open, E2E Passing, E2E Failing and Features Covered — followed by a 14-day open-findings trend, one bar per day showing how many findings were still open at the end of it. Below that sit three lists:
- Findings — each open one with its severity, the file it lives in, and how many times it has been seen, with three inline actions: Resolve, Dismiss and Reopen. A filter tabs between Open (n) and Resolved/Dismissed (n), so a finding you closed is still reachable rather than gone. Closing one fires straight from the row with no confirm step — it is fully reversible, and taxing every correct click to catch the rare wrong one would be the wrong trade — and a banner above the list carries an Undo for the finding you just changed, right where you are already looking.
- E2E Coverage — each feature with a PASS / FAIL mark.
- Least Recently Exercised — what the fleet has not touched in a while, or Never.
Everything refetches on a 30-second timer while the tab is open (the fleet produces results on a minutes-to-hours cadence, so a mount-only read would go stale the moment it landed), and there is a manual Refresh beside an "Updated <ago>" cue. Two different empty states say which problem you have: "QA fleet has not run on this machine yet" versus "QA fleet results could not be read" — a fleet that never ran is not the same as one whose records came back unreadable, and neither is dressed up as a reassuring zero.
Finished worktrees. There is no worktree cleanup dashboard. Finished worktree folders are
removed on their own by the worktree reaper
(Settings → Performance → Remove finished work areas automatically, on by default); its status
is GET /worktree-reaper.
Session-based cleanup (beta). A separate, cross-platform option that goes further than the free reaper: on an intelligent cadence it removes safe worktrees for you — including branches merged via a squash PR, which the plain reaper can't recognize — and, for anything it isn't sure about, spawns one short, conservative AI investigation that scans for stray worktrees and flags anything real to your inbox, then tidies itself away if it found nothing. It runs only on your active dev-pipeline repos (plus a slower sweep for stray worktrees anywhere), never deletes uncommitted or in-use work, and is bounded by a hard daily session + spend cap so it can't run up a bill — 5 sessions a day and $2 a day out of the box. Off by default; enable it and set the caps under All settings, where the session ceiling is adjustable from 1–50 and the dollar ceiling from $0–100. It runs alongside the free reaper, never replacing it.
The developer-facing reference — the project id, the gating settings, the mobile rules and the panel layout — is its own page for whoever is working on the panel.
Related
Dev Pipeline panel is the first half — the panel's overview, its Runs, Stages and Setup rail, and the Agent Board and Worktrees views inside it. Dev Pipeline documents the workflow these controls drive.
Last verified 2026-10-18