mesh-build-image's two tests that composed with catalogue.Resolve (gone with the composition) now hold the module's own
facts: its registry file is the placement and no file binds the store; it requires rootless-image-build, which buildah
provides and docker does not. What only proved composition is listed in .work/sdk1/proofs-for-node-engine.md. sdkjudge's
composition list drops the six refusals that left mesh-sdk.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The test copies the node-engine's repository from beside the check (MESH_CHECK_BESIDE, mesh-host, with mesh-sdk's Go module
beside it as one workspace when a group places it), has the module build github.com/novox/mesh-host/cmd/mesh-compose there,
and runs it; it fails loudly when the repository is not beside.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The module already builds whatever repository, directory and package the ask names; the tests now say so for the node-engine's
repository (a Go module at its root, the command mesh-compose) instead of mesh-sdk's composer, which leaves mesh-sdk, and the
real-bus test builds a stand-in command in its fetched tree. The README tells the runner's new home and what a build says at
its end.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The new argument files, a list of directories (the root as the empty name), adds files, {directory: {path relative to it:
blob id}}, every blob under each directory from the call's one recursive listing. A directory the commit lacks fails naming
it, a cut listing still fails whole, and a badly spelled directory is refused before the forge is asked.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The digest counted a module outside the tree by its version and checksum, so every mesh-sdk merge invalidated each kept
verdict of each repository. A package of the mesh's forge is now hashed by the files the tests reach, as a replacement to a
directory already was; any other outside module still counts by version and checksum. Every kept verdict is invalidated
once, by the form change.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
ModuleTrees answers each module directory's git tree, its manifest and lock digests, and the git objects of the paths its
artifacts name; it no longer computes delivery.NamedTree or reads the mesh packages out of go.sum.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
A build ask whose job built an artifact asks the new step build_go_imports once: `go list -deps` of each Go bundle's program,
no tests, the forge's packages outside the module's own directory (a sibling package of the same repository counts), sorted
and each once. The answer carries them as BuildAnswer.Imports. The step is also a tool, to read them by hand.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The runner, the toolchain guard, the fetch, the parsers, buildcpu, the conformance kit and mesh-pack move from mesh-sdk's
buildstep into modules/mesh-build-go/buildstep (own go.mod). mesh-build-go, mesh-build-node and mesh-build-image require it
by a same-repository replace; what the delivery parts read (the failure contract, the fetch request, the stale sweep, the
module tree shape, NewID, CapacitySetting) stays imported from mesh-sdk as sdkstep.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The node-engine alone composes (ADR 0593 §6, §8), so the provisioner stops
composing with frombus and fromruntime and reads each module's plan from
composed-<node> (read as mesh-host.composed): at its want, or the build its
have runs, and at the indexer's newest build, so a key ahead of its want
still carries its plan entry. A refused or not yet published composition
holds the module with its last plan kept. The golden test proves the plans
and provisions are today's, captured from the composing holder before the
change. It no longer reads the builds or the resolved values.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh/check pass 019614aedc14 on fe7c5ccb64bf | unit 6028 run, 142 kept, 16 skipped | integration 6091 run, 174 kept, 25 skipped | over bound: unit, in…
With the controller gone nothing made the pair row of a new consumer: the vault minted only for a provider
handed to the provisioner (ADR 0349 clause 17's premise, that a holder filled the old way gets the controller's
value). A node composing its own now asks the vault for its credentials (own, ADR 0547 §2, ADR 0550 §5), so the
vault makes the first value of a shared or per-login provision for every holder, create-only (issue 1126; ADR
0570 §6). A remake or a rotation still waits while a holder is filled the old way. A credential asked of the
vault in mesh-provisioner's provisions starts a pass at once, one pass at a time, and vault_status answers each
provider and provision's last outcome as `credentials`.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh/check checking fa6dc3a48b0c on 32c4c52261c4 | novox/mesh-catalog#594 in group feat/1126-removal
When the forge's listing of pull request heads fails, the fetch now says it
as a download failure that may be retaken, so a forge outage is asked again
and not held as a refusal (issue 1748, issue 1126 review 3).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The group's check has no name lookup, so go failed on the network before it
could say it knew no such revision, and the fetch, reading go's words, took
the asked head for a download failure. Whether a commit is a pull request's
head is now the forge's fact, asked whenever go cannot fetch it; go's words
only label a commit on no head (issue 1748, issue 1126 review 2).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh/delivery-group group feat/1126-removal rejected: composed together: the checker's verdict at the members' heads (a523066f1791): integration: meshbuildgo/c…
mesh/check failed: integration: meshbuildgo/cmd/mesh-build-go… | 7f8701637f9d on 32c4c52261c4 | judged by the candidate's mesh-build-go at b82f7f676986
nox cleanup delete without --confirm deletes nothing; the vault refuses it
for its missing confirm alone and judges only a confirmed delete, and nox
records the warranting act only for that one (issue 1126 review 2).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
nox records a hand act before each sensitive vault verb, with the verb's
address as its verb and cause operator; two in a fortnight raised a false
healer-wanted. mesh-healers reads those four verbs as decisions by design
(issue 1126 review 2).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
A reader on an older mesh-sdk refuses a provision field it does not know,
so a key a plan names carries no entry: such readers skip only a key ahead
of its consumer's want, which they could never bind (issue 1126 review 1).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh-provisioner keys provisions from a consumer's wanted and newest builds
but writes the wanted build's plan alone, so a requirement the newest build
adds was never bound. Each key now carries its plan entry, and a key ahead of
its want says in its why that no plan names it yet and its provider binds it
from the entry (issue 1126 review 1).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller is not placed (ADR 0590), yet the vault still answered a
caller named controller alone (accept), took its relay of who asked
(secret-ask, rotate's by) and sealed it the core's credentials. Those
branches and accept go with their tests; a node's account alone is never
minted or rotated here. The sensitive-verb judge now needs the act's verb to
be the vault's own, mesh-vault.<verb>, not any verb ending so (issue 1126
review 1).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
Nobody raises a lost or a far-behind consumer since the self-check's bus-consumer probe ended (issue 1126): the grants
part makes a missing account consumer itself and says one far behind as mesh-grants.consumer-behind, and H4's repair
was the controller's. The mesh defines no consumer of its own, so H-bus's consumer half judged nothing.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh-hand-acts records one what per act, so an approve of two consumers
could never be warranted. The judge reads each word of the act's what as a
name, and nox records one act listing the consumers (issue 1126 review 1).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
Each sensitive vault verb judges its own caller, and nox reaches the vault
as its machine's runtime, never as a person, so it could use none of them.
Before rotate, retire approve, cleanup delete and secret export nox records
the act through mesh-hand-acts (the vault verb as its verb, what it acts on
as its what) and names it as act; a refused recording calls the vault not at
all (ADR 0590 §8, issue 1126 review 1).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
With the controller ended mesh-sdk defines no mesh consumer, so the probe judged none and always passed (issue 1126).
The account consumers are the grants part's, which now says their differences and delays as conditions of its own;
the probe's three kinds are carried as ended in mesh-conditions' table.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The removal test took a grant written by hand; it now composes it with
BusRemovalGrants for controller-leftovers and the node-engine's acts, so
it proves the mesh's own grant is enough and no wider (issue 1126 B5).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
move-asks copied the controller's record of asks into the vault's once;
the controller is not placed (ADR 0590) and nothing else calls it. move-in
stays: the node-engine calls it. The drift test names the node's own
composition now that the controller composes for none (issue 1126).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
mesh-sdk drops compare, take-over and hand-back from the keeper's and
config's seats (ADR 0590): there is no controller to compare with or hand
back to. mesh-conditions loses those verbs, its take-over key and asks;
mesh-config reads its own sources always and pull-declares is gone; and
mesh-hand-acts no longer names the controller's log (issue 1126).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller judged the operator's account for approving or deleting a
retirement, rotating a secret and exporting sealed copies; it is not
placed, and these verbs are reached through the mesh MCP server on each
machine. Each now answers the operator's account (person.<name>), or a
caller naming its own recorded hand act for that verb and subject, and
nothing else (ADR 0590 §8, issue 1126).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller is not placed (ADR 0590), so its tools, modules, nodes and
node verbs answer nothing. claude-code now reads the seats' verbs from the
seats' contract, the placed modules' replaces from the indexer's manifests,
and the machines and where modules run from mesh-placement's placement; the
rendered instructions name the seats that own each job (issue 1126 B13).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller is not placed and its assignments mirror is frozen (ADR
0590). artifact-store-tools already answers through the seat's holder, so
its controller client goes. mesh-provisioner-client reads the seat
mesh-placement's placement, which its account is granted whole: a module
placed here stays while the seat holds no plan of it, filled from its last
plan, and leaves only when neither names it (issue 1126 B12).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
Both stand-in judges answered the fixed due date 2026-11-01; since 2026-11-02 00:00 UTC that date is past,
signalEntries drops the mail, no tell is sent, and both tests time out on every check that rebuilds the module
(met in the enrolment group's check of issue 774, 2026-11-02 02:50).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller is not placed (ADR 0590): messenger no longer treats it as
a condition owner or a walk's subject, mesh-watcher reads only where the bus
runs, restic tells the operator to compose the machine or set config through
the seats that do it now, and mesh-issues' examples name a placed repository.
mesh-issues keeps mesh-controller among its owners: issues located there,
issue 1126 among them, and route-proxy until it moves, still name it
(issue 1126 B11).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller ended by its unassignment, so its consumers, streams and
buckets may be removed as recorded hand acts (ADR 0590 §6). The nats tools'
account declares controller-leftovers in place of the lease write's two
words, and the removal test proves a consumer of EVENTS and a bucket's
stream go while EVENTS and its other consumers stay (issue 1126 B5).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
controllerdeps refuses any manifest naming the controller's seat, states or
buckets, the lease and the entries once kept to its write among them, and
any module file naming one of its buckets (ADR 0590 §5). The words check
refuses a controller address in any text a module shows. The handacts check
and noplacement's list of assignments readers are folded into it (issue 1126 B10).
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The vault deletes a retired holder only when confirm repeats its name, so
sending confirm "true" could never delete anything. The stand-in console now
also says which machine answered as a block of its own, as node-tools does.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The console appends which machine answered after the JSON, so the nodes
table decoded nothing and printed the raw answer; read the first value
and keep the trailer.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller ended by its unassignment, so its lease is never handed and
both stayed in their controller-era branch (issue 1126, A5). mesh-config no
longer trusts the controller's relay: a trusted value needs the operator's
Approve. The condition keeper carries nothing from the controller, clears
once at start what it carried, and refuses take-over as the one source.
Both drop their manifest read of mesh-controller.lease.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
nox now reaches the mesh through this machine's console, so the module
requires mcp-endpoint and writes the bound address to /etc/mesh-cli/console
rather than trusting a fixed port. No invokes: mesh-cli has no bus account,
and one nothing dials would only be minted.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller nox forwarded every line to is unassigned (issue 1126), so nox
answered nothing but enrol open. Each line now calls the seat verb or module
tool that owns its job through this machine's console (mesh_call); a line no
seat owns says so and names what to use, never failing silently.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller ended by its unassignment, so its lease key is absent for good
and the healers stayed in shadow while nobody swept the artifact store (issue
1126, A4). Drop both gates and each manifest's read of mesh-controller.lease;
the would bucket stays readable until its keys expire.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
The controller ended by its unassignment and its lease key expired, so the
gate on a handed lease kept retirement, rotation and node-login rotation off
for good (issue 1126, A3). Drop the gate, the lease watch and the manifest's
read of mesh-controller.lease; vault_status no longer reports a lease.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
jschoubben
mesh-tools