• Joined on 2026-03-08
ltms commented on pull request fleet/fleetd#462 2026-09-10 14:20:08 +02:00
fleetd #439: omit fleet_list's coordinator key for non-primary callers

Merged as 92a96fc. Closing by hand — a local --no-ff merge plus push did not trip the forge's auto-close.

Proof the branch is in main: `git merge-base --is-ancestor origin/worker/439-co…

ltms commented on pull request fleet/fleetd#468 2026-09-10 14:20:03 +02:00
fleetd #464: guard charter tool names

Merged as 49df792. Closing by hand — a local --no-ff merge plus push did not trip the forge's auto-close.

Proof the branch is in main: `git merge-base --is-ancestor origin/worker/464-ch…

ltms commented on pull request fleet/fleetd#465 2026-09-10 14:20:00 +02:00
fleetd #458: restate invariant 5 by purpose, not mechanism

Merged as eccd054. Closing by hand — a local --no-ff merge plus push did not trip the forge's auto-close.

Proof the branch is in main: `git merge-base --is-ancestor origin/worker/458-in…

ltms commented on pull request fleet/fleetd#467 2026-09-10 14:19:57 +02:00
fleetd #463: default listFleet callerIsPrimary to false, fail closed

Merged as 235644c. Closing by hand — a local --no-ff merge plus push did not trip the forge's auto-close.

Proof the branch is in main: `git merge-base --is-ancestor origin/worker/463-li…

ltms commented on pull request fleet/fleetd#457 2026-09-10 14:19:52 +02:00
fleetd #446: make exhaustedPattern hot, name the fix, report it in fleet_profiles

Merged as 1fb6176. Closing this PR by hand: the merge was made locally with --no-ff and pushed, which did not trip the forge's auto-close.

Proof the branch is in main, not just…

ltms closed issue fleet/fleetd#463 2026-09-10 14:17:13 +02:00
listFleet's compat overloads default callerIsPrimary to true, so the coordinator gate fails open
ltms closed issue fleet/fleetd#446 2026-09-10 14:17:10 +02:00
The model gate can be turned off at runtime but its usage-limit detector cannot be turned on at runtime
ltms commented on issue fleet/fleetd#460 2026-09-10 14:17:04 +02:00
The /mcp transport has no test: 57 endpoint tests all enter through REST, the door no agent uses

A concrete, cheap case for this ticket, measured today

#446 round 3 (merged 1fb6176) extracted the quarantine ExhaustionSink out of Fleetd.main into a static factory, `Fleetd.exhaustionSi…

ltms commented on issue fleet/fleetd#446 2026-09-10 14:16:52 +02:00
The model gate can be turned off at runtime but its usage-limit detector cannot be turned on at runtime

Merged as 1fb6176 (PR #457), after three rounds.

mvn -B clean test on the merged tree 953ce11: Tests run: 1601, Failures: 0, Errors: 0, Skipped: 0, BUILD SUCCESS, 0 compile-error…

ltms pushed to main at fleet/fleetd 2026-09-10 14:16:29 +02:00
1fb6176783 Merge #457: exhaustedPattern goes hot, and the warning names the fix (fleetd #446)
7772b41993 fleetd #446 round 3: extract exhaustionSink and pin its caller (Cell A/B)
30d6872779 fleetd #446 follow-up: pin the WARNING text and the fleet_profiles model/reason fields
45aca9eb3e fleetd #446: make exhaustedPattern hot, name the fix in the warning, report it in fleet_profiles
Compare 4 commits »
ltms opened issue fleet/fleetd#469 2026-09-10 14:13:11 +02:00
Charter text is still never checked against the live tool surface — one canonical tool-name set (follow-up to #464)
ltms commented on issue fleet/fleetd#464 2026-09-10 14:12:30 +02:00
Nothing checks charter text against the tool surface the server registers

Correcting one number in the comment above

I wrote that FleetConfig.validateCharters() has "0 lines of its body mention a tool name". **The conclusion is right; the measurement behind it was…

ltms commented on issue fleet/fleetd#464 2026-09-10 14:11:25 +02:00
Nothing checks charter text against the tool surface the server registers

Merged as 49df792. One new test file, 72 lines, no production change.

CharterToolSurfaceTest pulls every fleet_* / bridge_* token out of the configured charters and every `tool("fleet_…")…

ltms pushed to main at fleet/fleetd 2026-09-10 14:11:06 +02:00
49df79203c Merge #464: a test that charter text names only registered tools (fleetd #464)
c4e23eebad fleetd #464: guard charter tool names
Compare 2 commits »
ltms commented on issue fleet/fleetd#455 2026-09-10 14:07:30 +02:00
Invariant 5 ("never drive the multiplexer directly, no socket") is unsatisfiable for a worker assigned to herdr code

Criterion 4 decision: the addendum note stays

#458 shipped as eccd054. Canonical invariant 5 now reads:

  1. Never move a fleet session, pane or peer except through the bridge. The…
ltms commented on issue fleet/fleetd#463 2026-09-10 14:07:20 +02:00
listFleet's compat overloads default callerIsPrimary to true, so the coordinator gate fails open

Merged as 235644c (PR #467). Main source change is one word plus javadoc.

 return listFleet(workers, sessions, messages, capacity, healthCoverage, quarantine, outage,
-       …
ltms pushed to main at fleet/fleetd 2026-09-10 14:05:02 +02:00
235644c0f0 Merge #467: listFleet's callerIsPrimary default fails closed (fleetd #463)
7df503dfc2 fleetd #463: default listFleet's callerIsPrimary to false, fail closed
Compare 2 commits »
ltms closed issue fleet/fleetd#458 2026-09-10 13:59:23 +02:00
Canonical invariant 5 states a mechanism where it means a purpose — restate it (blocked on #455)
ltms commented on issue fleet/fleetd#458 2026-09-10 13:59:21 +02:00
Canonical invariant 5 states a mechanism where it means a purpose — restate it (blocked on #455)

Merged as eccd054 (PR #465). One file, one hunk.

Invariant 5 now reads:

  1. Never move a fleet session, pane or peer except through the bridge. The bridge owns policy; the multiplexer…
ltms pushed to main at fleet/fleetd 2026-09-10 13:57:30 +02:00
eccd0548ce Merge #465: state canonical invariant 5 as a purpose, not a banned tool (fleetd #458)
29e7a06c49 fleetd #458: restate invariant 5 by purpose, not mechanism
Compare 2 commits »