* Count review rounds per pull request across /ship runs (GST-2673) /ship's two-round review cap was counted inside each run, so a pull request shipped many times was reviewed again at every head. Step 1 now reads the rounds the PR body's receipt history records, passes them to the review plan, and sends later heads to the scoped re-check, with a lane or Captain round grant for an extra round. Adds the no-loss replay preregistration. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Cap full review rounds per pull request and keep fix-checks to each run The plan counts a PR's earlier full rounds (--prior-full) against the cap of two, as the approved plan says, and prints every earlier round as pr_rounds=. Renames the troubleshooting anchor to pr-round-cap, lists <SESSION_ID> as a reviewed placeholder, and leaves ship/DRIVER.md at its size cap. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Point the ship driver's review pass at Step 9's plan flags, which carry the PR round count Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Give a pull request's final head a full review and make the diff-only re-check advice only (GST-2673) A scoped re-check that reads only what changed since the last reviewed head caught 22 of 24 planted faults but none of 8 real faults that full rounds had caught, so it no longer carries a /ship receipt. A later head whose own lines changed now gets a fresh full review, the PR's last before merge; a PR already at two full rounds gets one read-only full review with no fix round, whose blocker goes to the lane's Captain. carry prints RECHECK: advisory and refuses --recheck. The drift threshold, the re-check blocker count and the design-change grant kind are removed with the route they fed. Repins lib/review-depth.ts and ship/sections/review-army.md under replay GST-2673 (docs/evals/pr-round-count-results.md; deterministic grid against main: 196,608 of 196,608 plans identical with no prior rounds). Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Keep a review round grant usable through its fix round and after a read-only review A grant now holds after the fix round's commits move HEAD past it, counts only the rounds earlier heads spent, and a receipt written without a run keeps other runs' stubs. Review-Finding: p1/specialist-batched#1 reader=state-machines Review-Finding: p1/red-team#2 reader=red-team Review-Finding: p1/red-team#3 reader=red-team Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Count every review round at a granted head again, so a grant can't reset the count Reverts the history change that skipped rounds recorded at a round grant's head: the run that hit the cap is recorded at that same head, so skipping it wiped the count. A grant recorded after a read-only review still doesn't buy a fix round (fails closed). Review-Finding: p2/adversarial-native#1 reader=adversarial-native Review-Finding: p2/outside-voice#1 reader=outside-voice Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> --------- Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
gstack on Capy
This folder holds the parts of gstack that only Capy crews use: the tools that
run a crew of coding agents under a Captain. It stays in the public repo, but
./setup loads it only on a Capy machine. Anywhere else, setup skips it and
gstack works exactly as it does without it.
What Capy users get on top of gstack
- Captains.
/captainruns many agent threads from one controller: lanes send a READY block and never merge, the controller keeps the digest, relays approvals in a fixed shape and runs a daily flaky-test pass (docs/captain.md). - A merge queue.
gstack-ship-queuelands verified pull requests alone or in trains, checks the exact head before it merges, and stamps the version only at release time. - Receipts.
/shipleaves a receipt for the head it checked, so the controller can tell that a pull request was reviewed and tested at the commit it merges. - Budgets.
/shipand/autoplanmeter their own spend against the cap a brief sets, and stop or hand off before they cross it (docs/ship-budget.md).
How the whole factory fits together: capy/README.md.
How it is loaded
A Capy machine is one where CAPY_MACHINE_CONFIG is set (cloud machines) or
~/.capy exists (cloud machines and paired devices, such as a Mac running
Capy). Any ~/.capy counts, so when it is unclear the crew skills install
rather than a Capy user losing /captain. gstack_on_capy in
bin/gstack-install-registry.sh is that test, and
GSTACK_CREW_SKILLS there lists the crew skills. On a Capy machine setup and
gstack-relink register them like any other skill; elsewhere they get no entry
on any host, and an entry an earlier install made is removed.
On Capy the session launch file (bin/gstack-capy-install) also puts
crew/bin/ on PATH, so the crew commands keep their bare names there. Public
CI runs crew/test/ with the rest of the free suite.
What is here
| Path | What it holds |
|---|---|
bin/ |
gstack-release-gate (the paid gate on main every 8 hours) and gstack-coauthor-trailers (credit for contributor lines a rewrite carries over) |
lib/ |
merge trains and waves (ship-queue-train*, ship-queue-wave, the land lock, screens, rechecks), the release gate, version-gate trains and lane ownership |
scripts/ |
the release cut and its daily check (release-cut, release-cadence), and the paid-eval CI steps only our crews run (train gate and backstop, gate cost, gate reuse, receipt recovery, census arms, judge calibration configs) |
test/ |
their tests |
docs/ |
the manuals: running a Captain crew and the recommended harness |
The crew tooling moves here in stages. Code that no public skill calls comes
first; the steps /ship, /review and /autoplan take only for crews move
after they call it through one switch. Some crew tools stay outside crew/ for
now because public code still calls them: /captain's own folder (the skill
generator reads top-level folders), gstack-ship-queue, receipts, the
pre-gate, spend caps and approvals.