@staix/agent-hub 0.8.0 → 0.8.1

package/CHANGELOG.md

@@ -2,6 +2,11 @@
 
 Issue and pull request numbers in the entries for 0.7.7 and earlier refer to the previous repository, archived on 2026-09-30 when this repository's history was rewritten; the one exception is the open smoke-check issue, formerly #12, which moved here as #1. Numbers in newer entries refer to this repository.
 
+## 0.8.1
+
+- Upgrading a running 0.7.x hub that has a task on its board no longer stops at verification ("queue, manual pause, task board or budget preservation was not verified"): the 0.8.0 target digested its tasks with the new `plan` field, which the 0.7.x source never had. The target now also accepts the digest in the source's shape while every plan is empty; any real change to the board still fails the check. Use the 0.8.1 coordinator, not 0.8.0's, for such an upgrade (#57).
+- `ahub kill` returns only once the hub has released its registry claim, so an `ahub projects remove` right after it no longer refuses, and an `ahub up` right after it no longer reads the project as starting and starts nothing (#58).
+
 ## 0.8.0
 
 - Plans and completed-change notices: `hub_task_propose` and `hub_task_accept` take a `plan` (paths, symbols, signatures, insertion points), overlaps also count plan paths and shared symbols, the owners of overlapping tasks get the plan as a ride-along line, and when a task is done they get a notice of the changed files, signatures and summary. PII tasks are left out on both sides, and a plan matching a PII pattern makes a new task a PII task or is refused on an ordinary one. Overlap mentions, ride-alongs, completed-change notices and overlap events leave out any name that matches a PII pattern, and model-written titles, refs and plan items are folded onto one line. During a PII turn the local worker can no longer finish, review or accept-with-plan an ordinary task, which would carry its words to other peers and claude-mem (#31).

package/README.md

@@ -4,7 +4,7 @@ Native multi-agent hub for one developer's machine: Claude Code, Codex, Kimi Cod
 hub-owned local-LLM worker collaborate as peers in independent project directories, with
 task-aware model routing (Switchyard) in front of a self-hosted gateway (OmniRoute).
 
-Status: 0.8.0, control protocol 10. Durable delivery records distinguish queued
+Status: 0.8.1, control protocol 10. Durable delivery records distinguish queued
 work from uncertain execution. The [smoke checklist](docs/smoke.md) records
 verified paths and remaining prerequisites.
 
@@ -28,7 +28,7 @@ cd <your project> && ahub init && ahub up && ahub tail
 Or install the same version from GitHub:
 
 ```bash
-bun add -g github:STAIxBWLB/agent-hub#v0.8.0 && ahub setup
+bun add -g github:STAIxBWLB/agent-hub#v0.8.1 && ahub setup
 ```
 
 The installed commands remain `ahub` and `agent-hub`.

package/docs/operations.md

@@ -1,6 +1,6 @@
 # Operations guide
 
-This guide describes ahub 0.8.0 and control protocol 10. Live verification
+This guide describes ahub 0.8.1 and control protocol 10. Live verification
 results and remaining prerequisites are recorded separately in [the smoke ledger](smoke.md).
 
 ## Install and start
@@ -302,8 +302,8 @@ source and carries every recovery fix released up to it. Protocol 8 and older
 project directory, without replacing the global CLI first:
 
 ```bash
-bunx --package @staix/agent-hub@0.8.0 ahub upgrade --to 0.8.0 --dry-run
-bunx --package @staix/agent-hub@0.8.0 ahub upgrade --to 0.8.0 --yes
+bunx --package @staix/agent-hub@0.8.1 ahub upgrade --to 0.8.1 --dry-run
+bunx --package @staix/agent-hub@0.8.1 ahub upgrade --to 0.8.1 --yes
 ```
 
 | Running now | Coordinator to use |
@@ -313,7 +313,12 @@ bunx --package @staix/agent-hub@0.8.0 ahub upgrade --to 0.8.0 --yes
 | any supported source, with the installed CLI already at the target | `ahub upgrade` below, which is the same coordinator |
 | 0.5.x or earlier (protocol 8 and older) | not supported: bootstrap by hand with the matching CLI |
 
-Do not use an older installed CLI as the coordinator. A 0.6.x CLI cannot target
+Do not use the 0.8.0 coordinator for a running 0.7.x hub with tasks on its board:
+its verification never matches the board and the operation stays blocked (fixed
+in 0.8.1). Such a blocked operation can neither resume nor abort, and its lock
+refuses `up` and `kill` for every project; the [smoke ledger](smoke.md) (0.7.11
+to 0.8.0) records the manual cleanup. Do not use an older installed CLI as the
+coordinator. A 0.6.x CLI cannot target
 protocol 10: its plan does not check the target's protocol, so the dry-run shows
 no blocker, and `--yes` stops at staging ("target protocol requires a newer
 coordinator") with an operation left to clear by `ahub recovery abort <id>`. An
@@ -325,19 +330,19 @@ The coordinator verifies and retains the exact target package, preserves its
 own source, and promotes the global CLI only after restored projects pass
 readback.
 
-Once the installed CLI is 0.8.0, review the current project or all registered
+Once the installed CLI is 0.8.1, review the current project or all registered
 projects first:
 
 ```bash
 ahub restart --dry-run
-ahub upgrade --to 0.8.0 --dry-run
+ahub upgrade --to 0.8.1 --dry-run
 ```
 
 Apply only after reviewing the plan:
 
 ```bash
 ahub restart --yes
-ahub upgrade --to 0.8.0 --yes
+ahub upgrade --to 0.8.1 --yes
 ahub recovery status <operation-id>
 ahub recovery resume <operation-id>
 ahub recovery abort <operation-id>

package/docs/quickstart.md

@@ -17,7 +17,7 @@ ahub setup                                # installs the Claude Code channel plu
 Or use the matching GitHub release:
 
 ```bash
-bun add -g github:STAIxBWLB/agent-hub#v0.8.0
+bun add -g github:STAIxBWLB/agent-hub#v0.8.1
 ahub setup
 ```
 

package/docs/smoke.md

@@ -859,3 +859,27 @@ the claim overlap warnings (0.7.6 and later), then the owner decides in issue #8
   and that Codex's running session no longer knows the turn (ask it about the
   reverted turn): the schema says `thread/revert` changes only the saved history.
 
+## 0.7.11 to 0.8.0 attended upgrade (2026-10-01)
+
+A scratch project (a git work tree whose `.agenthub/config.json` predates 0.8.0
+and has no `snapshots` block) ran a 0.7.11 hub with one proposed task and no
+peers attached.
+
+- `bunx --package @staix/agent-hub@0.8.0 ahub upgrade --to 0.8.0 --dry-run`
+  exited 0: one project, source 0.7.11, protocol 10, no blockers.
+- `--yes` prepared and committed the source, started the 0.8.0 target and
+  reached the restored-peers phase (no peers were attached), then stopped at
+  verification: `phase: blocked`, "queue,
+  manual pause, task board or budget preservation was not verified". The
+  global CLI stayed at 0.7.11.
+- Cause: the 0.7.11 source digested its task rows without `plan`, and the
+  0.8.0 target's rows carry `plan: {}` after the #31 migration, so any board
+  with a task failed the check. Fixed in 0.8.1 (#57).
+- The blocked operation could not resume (the same target fails the same
+  check) or abort (the source was already stopped), and its recovery lock
+  refused `up` and `kill` for every project. The scratch hub was stopped under
+  the operation's environment, and the operation was marked cancelled with its
+  lock released: an operator step, not a designed path.
+- The release gate for v0.8.0 (Actions run 36807880478) failed once on
+  "removing a running project is refused"; the rerun passed and published. Cause: `ahub kill` returned before
+  the hub released its registry claim. Fixed in 0.8.1 (#58).

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@staix/agent-hub",
-  "version": "0.8.0",
+  "version": "0.8.1",
   "description": "Native multi-agent hub: Claude Code, Codex, Kimi Code, Pi and local inference as peers in one project",
   "license": "MIT",
   "type": "module",

package/plugins/agent-hub/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "agent-hub",
-  "version": "0.8.0",
+  "version": "0.8.1",
   "description": "Channel between Claude Code and the agent-hub daemon: peer messages from Codex, Kimi and the local worker arrive as channel events; hub_send replies.",
   "author": {
     "name": "Young Joon Lee",

package/plugins/agent-hub/server.js

@@ -15727,7 +15727,7 @@ class ControlClient {
 // package.json
 var package_default = {
   name: "@staix/agent-hub",
-  version: "0.8.0",
+  version: "0.8.1",
   description: "Native multi-agent hub: Claude Code, Codex, Kimi Code, Pi and local inference as peers in one project",
   license: "MIT",
   type: "module",

package/src/hub/daemon.ts

@@ -673,12 +673,22 @@ export async function startDaemon(opts: DaemonOptions) {
     return [id, row];
   }));
   const digest = (value: unknown) => createHash("sha256").update(JSON.stringify(value)).digest("hex");
-  const integrity = () => {
+  // `prePlan`: the board as a hub from before the plan column (#31, 0.8.0) digested it. An empty plan is left out, so an
+  // upgrade from such a hub still verifies the board it was handed; a real plan never matches that shape.
+  const integrity = (prePlan = false) => {
     const queues = Object.fromEntries(Object.keys(bus.snapshot().queues).sort().map((id) => [id, bus.queueIds(id)]));
-    const boardState = board.list().sort((a, b) => a.id - b.id);
+    const tasks = board.list().sort((a, b) => a.id - b.id);
+    const boardState = prePlan ? tasks.map(({ plan, ...t }) => (plan && Object.keys(plan).length ? { ...t, plan } : t)) : tasks;
     const budgetState = budget.persistedPauseDigestRows().sort((a, b) => a.peer.localeCompare(b.peer));
     return { queues, manualPaused: [...manualPaused].sort(), boardDigest: digest(boardState), budgetDigest: digest(budgetState) };
   };
+  /** The integrity in the shape `expected` was recorded in: the current one unless only the pre-plan shape matches. */
+  const integrityAs = (expected: unknown) => {
+    const now = integrity();
+    if (!expected || JSON.stringify(expected) === JSON.stringify(now)) return now;
+    const old = integrity(true);
+    return JSON.stringify(expected) === JSON.stringify(old) ? old : now;
+  };
   const status = () => ({
     projectId,
     instanceId,
@@ -1200,7 +1210,7 @@ export async function startDaemon(opts: DaemonOptions) {
       ...([...bus.peers].filter(([, peer]) => peer.state === "busy").map(([id]) => `${id} is busy`)),
       ...(permissions.size ? ["pending approvals"] : []),
     ],
-    integrity: { current: integrity(), ...(restored?.integrity ? { expected: restored.integrity } : {}) },
+    integrity: { current: integrityAs(restored?.integrity), ...(restored?.integrity ? { expected: restored.integrity } : {}) },
     peers: Object.fromEntries([...(recoveryPeerSnapshot ?? []), ...Object.values(recoveryPeers()).filter((peer) => !(recoveryPeerSnapshot ?? []).some((saved) => saved.id === peer.id))].map((peer) => {
       const now = recoveryPeers()[peer.id];
       if (recoveryPhase === "restored" || recoveryPhase === "released") return [peer.id, now ?? { id: peer.id, state: "offline", queueIds: bus.queueIds(peer.id) }];
@@ -1298,7 +1308,7 @@ export async function startDaemon(opts: DaemonOptions) {
       const missing = (recoveryPeerSnapshot ?? []).filter((saved) => saved.state !== "offline" && !manualPaused.has(saved.id) && !budget.record(saved.id) && (!current[saved.id] || current[saved.id]!.state === "offline"));
       if (missing.length) return recoveryError(`required peers are not attached: ${missing.map((peer) => peer.id).join(", ")}`);
       if (!recoveryReady()) return recoveryError("required peers or approvals are not ready");
-      if (!restored?.integrity || JSON.stringify(restored.integrity) !== JSON.stringify(integrity())) {
+      if (!restored?.integrity || JSON.stringify(restored.integrity) !== JSON.stringify(integrityAs(restored.integrity))) {
         return recoveryError("queue, pause, task board or budget integrity changed during recovery");
       }
       archiveRestartSnapshot(opts.stateDir, op);

package/src/hub/lifecycle.ts

@@ -179,6 +179,15 @@ export async function startProject(project: Project, options: { unattended?: boo
   } finally { registry.close(); }
 }
 
+/** Whether `instanceId` still holds the project's registry claim, counted as held while its process may be alive. */
+function claimHeld(id: string, instanceId: string): boolean {
+  const registry = new Registry();
+  try {
+    const row = registry.get(id);
+    return row?.instanceId === instanceId && (row.pid === null || processAlive(row.pid) !== false);
+  } finally { registry.close(); }
+}
+
 export async function stopProject(project: Project, expectedInstance?: string): Promise<void> {
   assertLifecycleAvailable();
   const inspection = await inspectProject(project);
@@ -194,8 +203,10 @@ export async function stopProject(project: Project, expectedInstance?: string):
   const deadline = Date.now() + 10_000;
   while (Date.now() < deadline) {
     const current = readControl(project.stateDir);
-    if (!current) return; // daemon removes its manifest only after owned children stop
-    if (current.instanceId !== instanceId) throw new Error("a new hub instance started; it was not stopped");
+    // The daemon removes its manifest only after owned children stop; its process releases the registry claim just
+    // after that. Until then `projects remove` refuses and `up` reads the project as starting, so wait for both.
+    if (!current && !claimHeld(project.id, instanceId)) return;
+    if (current && current.instanceId !== instanceId) throw new Error("a new hub instance started; it was not stopped");
     await Bun.sleep(100);
   }
   throw new Error("shutdown is still pending; state and ownership were retained");