@awak-app/simy-cli 0.7.35 → 0.7.37

package/README.md

@@ -8,7 +8,7 @@ This signs the desktop into SIMY; browser automation uses its own provider conne
 
 Technical coding failures retain their task and retry after a persisted 5–30 minute
 cooldown. Automatic recovery preserves the original goal, usage limits, provider
-connection and explicit stops. Enabling automatic continuation can also adopt an
+connection and explicit stops. Turning on Delivery Loop can also adopt an
 ordinary loop waiting for an external dependency, starting with read-only inspection.
 A damaged provider-link file can be repaired by **Connect SIMY** for that provider;
 the original file is retained and the other provider stays disconnected.
@@ -36,7 +36,7 @@ for configuration, tests and remaining host integration requirements.
 
 ローカルDelivery Loopの0.6.24候補を版固定で実行するには、[ローカル候補版のインストール・実行](docs/local-delivery-runtime.md)を参照してください。
 
-Local execution agent for SIMY Local Tasks and Agentic Loop runs.
+Local execution agent for SIMY Local Tasks and Delivery Loop runs.
 
 To supervise an explicitly requested current goal after a finished Codex run, use
 `simy delivery followup --provider codex --session SESSION_ID --previous-run-id RUN_ID --objective "Current goal and expected result"`.
@@ -75,7 +75,7 @@ segments, and keep lifetime failure counts after event rotation.
 
 Each AI app card separates **SIMY connection: on / stopped** from CLI, sign-in
 and plugin health. Disconnecting collapses those details and explains that SIMY
-monitoring and automatic continuation are stopped. Details can still be opened;
+monitoring and Delivery Loop are stopped. Details can still be opened;
 reconnecting restores the item statuses. Plugin inspection failure never changes
 the connection setting or hides its disconnect action.
 
@@ -116,7 +116,7 @@ The server's durable recovery migration requeues the same failed job and retains
 its input, history and attempt count; it is required to recover jobs previously
 held by the server's lifetime retry cap.
 
-Agentic Loop preflight accepts the task's existing requirement, acceptance
+Delivery Loop preflight accepts the task's existing requirement, acceptance
 criteria, required checks and charter context alongside the repository/provider.
 Its `startup` summary lists all blockers, unverified conditions and recovery
 actions together. Environment-only calls remain supported. Task checks reuse the
@@ -135,7 +135,7 @@ remain available for history and usage warnings. The Web controller supports
 connected local Codex/Claude engines; unsupported cloud execution is refused
 before dispatch.
 
-A completed Agentic worker's structured result is saved privately before evidence
+A completed Delivery Loop worker's structured result is saved privately before evidence
 collection and the final observation flush. If the controller restarts after that save,
 the authorized daemon recovers the same attempt and collects fresh evidence
 without dispatching the completed worker again. Collector read failures retry up
@@ -157,14 +157,14 @@ repeated on every monitor tick; explicitly continue the same run to recover.
 Older reviews without a saved CI basis can refresh only when their prior evidence
 still establishes the complete reviewed identity; otherwise explicitly continue.
 
-For interrupted Agentic work, `GET /v1/agentic-loop/:id/recovery` provides a
+For interrupted Delivery Loop work, `GET /v1/agentic-loop/:id/recovery` provides a
 read-only objective and checkout fingerprint. Reconcile an uncertain side effect
 with `POST` to that endpoint (`action: reconcile`, authoritative
 `operation_status`, `observation`, `evidence_refs`) before resuming. A paused
-Agentic run retains `POST /v1/agentic-loop/:id/continuation-ceilings` for an
+Delivery Loop run retains `POST /v1/agentic-loop/:id/continuation-ceilings` for an
 explicit safety-ceiling extension under the progress policy. Token and retry
 budgets are notification thresholds: they do not interrupt or stop useful work.
-Agentic Loop checks a separate safety ceiling before each new provider call:
+Delivery Loop checks a separate safety ceiling before each new provider call:
 15,000,000 weighted Provider tokens and 50 executor attempts by default, using
 existing `provider_token_ceiling` and `attempt_ceiling` settings when present.
 The final attempt may still be reviewed if token allowance remains. Active
@@ -175,7 +175,7 @@ remaining work, explicitly extend the relevant ceiling within the existing
 run. A notification-threshold increase alone cannot clear a safety stop; at the
 maximum ceiling, further automatic calls remain blocked. Extension itself never
 resets usage or resumes work. Local Task and managed Delivery behavior is not
-changed by this Agentic safety gate. Both routes require the original authorized
+changed by this Delivery Loop safety gate. Both routes require the original authorized
 account, organization and device throughout the request.
 
 ## Provider reconnection and saved consent
@@ -244,9 +244,9 @@ after reconnecting permits another probe, with bounded automatic cooldowns.
 Conversation checks blocked by **Sign-in required** automatically recheck after
 a fresh Codex or Claude login probe succeeds, including after an app restart.
 An unchanged ready status cannot repeatedly re-run a failed check. Manual retry
-remains available; continuing the conversation still requires **Auto-continue**.
+remains available; continuing the conversation still requires **Continue automatically**.
 
-You can enable **Auto-continue** while a conversation is running when its original
+You can turn on **Continue automatically** while a conversation is running when its original
 provider configuration is verified. SIMY saves the preference and waits for the
 current turn to finish before checking remaining work. Configuration and runtime
 blockers appear in the conversation view with recovery guidance; turning the
@@ -276,16 +276,18 @@ Closing the window keeps the agent running. Open the app again or choose
 interrupting active tasks. The normal `simy` command remains available.
 
 The macOS **Automatic updates** switch saves this Mac's choice across restarts.
-Compatible signed installations enable it by default. Patch/minor updates wait
-for idle and verify the new runtime before handoff; failed startup restores the
-previous installed version. The app then switches to the verified new interface
-when no unfinished work or confirmation is pending. Turning the switch off keeps
+Compatible signed installations enable it by default. Patch/minor updates prepare
+and verify the new runtime before a safe handoff; failed startup restores the
+previous installed version. Choose **Update and restart** to open the verified
+new interface, or **Later** to keep using the current window. The window never
+restarts automatically. Turning the switch off keeps
 manual checks and **Download DMG** available. An explicit `SIMY_AUTO_UPDATE=0`
 remains authoritative, and major updates continue to use the DMG.
 **Stop task…** in a task
 detail asks for confirmation and terminates its local Provider process, keeping
-existing output and history. An explicit update can likewise stop active work
-after confirmation; automatic updates still wait for idle. **⌘Q** exits SIMY and
+existing output and history. Window updates preserve the same execution owner,
+pending approvals and inputs, and scoped drafts. Older services without this
+capability keep running and wait for a safe update. **⌘Q** exits SIMY and
 its background services directly, including when work is running. The window shows
 the app version, and normal app launches select the newer installed interface.
 
@@ -314,7 +316,7 @@ missing sources. Provider plugin enablement and hook trust remain provider-owned
 
 For conversation continuation, select SIMY in the Codex app with `@` and ask
 `continue this task` (Japanese: `この作業を続けて`). The `continue` skill uses the
-running app's existing Auto Continue switch and background coordinator. In
+running app's existing Delivery Loop switch (**Continue automatically**) and background coordinator. In
 Codex CLI, use its native `$simy:continue` skill selector; in Claude Code use the
 native `/simy:continue` command. Claude Code does not define `@simy` as a plugin
 command: use the slash command for an explicit invocation. Merely mentioning
@@ -347,7 +349,7 @@ SIMY**, reopen it, and reconnect; closing the window alone does not restart the
 background host. Refresh connections and confirm the conversation list returns.
 Diagnostic `provider connection failed` entries distinguish `reason=update`,
 `organization_change`, and `shutdown` without including credentials or transcripts.
-Reconnection does not sign in to Codex CLI or restore revoked automatic continuation.
+Reconnection does not sign in to Codex CLI or restore a revoked Delivery Loop.
 
 The plugin also includes a non-blocking connection reminder. At task startup,
 resume, and prompt submission it reads the origin-scoped local discovery record
@@ -368,7 +370,7 @@ from the same maintained source as Codex; only provider terminology and hook/man
 format differ. Installing source files does not imply that either provider has enabled
 the plugin. Existing Claude settings, unrelated marketplaces, and plugins are preserved.
 
-State-only feature-improvement collection ships with the generated Claude plugin, but uploads require the paired desktop CLI runtime to refresh device consent and drain its local outbox. Without that paired authenticated runtime, the plugin remains default-off and does not enable Auto-continue.
+State-only feature-improvement collection ships with the generated Claude plugin, but uploads require the paired desktop CLI runtime to refresh device consent and drain its local outbox. Without that paired authenticated runtime, the plugin remains default-off and does not turn on Delivery Loop.
 
 After that first installation, the daemon checks installed workflows at startup
 and every hour, including after a CLI automatic upgrade. It updates only existing
@@ -379,7 +381,7 @@ marketplace still points to SIMY’s generated local source. Plugin versions inc
 receive a fresh cache. Use a new Codex task to pick up updated skills; running
 tasks are not restarted. Missing or disabled plugins are not enabled automatically.
 Failed refreshes retry on the next check without blocking daemon startup.
-Auto-continue ON saves consent for the selected original conversation; it does not
+**Continue automatically** ON saves consent for the selected original conversation; it does not
 immediately launch a second turn. A verified desktop discovery admission or an
 existing exact admission saves this preference without another provider RPC.
 After the current turn settles, continuation separately rechecks context, the
@@ -399,7 +401,7 @@ probes remain limited to recent conversations and explicitly retained work;
 older rows remain visible with unverified execution state. The macOS work list then prioritizes verified running
 conversations, including families with running children, followed by work needing
 a human response and recent activity. Its All, Running, Needs attention, Completed
-and Archived views combine with search. Tracking and auto-continue settings never
+and Archived views combine with search. Tracking and Delivery Loop settings never
 prove that an AI is running; observations older than 45 seconds or a failed
 provider refresh display an unknown execution state.
 
@@ -415,7 +417,7 @@ unfinished children or open PRs keep a completed task out of the Completed view.
 Codex candidates come from CLI, VS Code and app-server
 conversation sources; non-interactive exec, unknown sources and delegated agents
 are excluded. A hook observation alone cannot restore a filtered conversation.
-Explicitly tracked, monitored, auto-continue and unsettled supervision records
+Explicitly tracked, monitored, Delivery Loop and unsettled supervision records
 remain visible even during a provider discovery failure. When a retained Codex conversation is
 omitted from the recent provider page, SIMY reads its exact metadata without
 loading turns or taking writer ownership. The same interactive source filter and
@@ -475,7 +477,7 @@ settings can differ, and project-scoped installations are labeled explicitly.
 missing sources; existing sources still receive updates.
 
 This is the recommended installation because the CLI can safely install patch
-and minor releases, and the daemon restarts only when no Agentic Loop task is active. To run
+and minor releases, and the daemon restarts only when no Delivery Loop task is active. To run
 the interactive chat after installation, use `simy`.
 
 For a one-time evaluation or source development, these modes remain supported:
@@ -501,7 +503,7 @@ injected runtime context. Dated messages outside the configured lookback are
 excluded even when their conversation continues today. Legacy transcripts with
 no message timestamps use the file timestamp as a fallback. The local cursor stores only opaque hashes and
 timestamps scoped to the connection, account, and device. Uploads require the
-`user_conversation.v2` Web, Crew, and Backend contract and exact per-chunk
+SIMY service `user_conversation.v2` contract and exact per-chunk
 receipts before the cursor advances. Deploy those companion changes before
 enabling this CLI version; earlier receivers cannot acknowledge v2 history.
 
@@ -528,18 +530,18 @@ cursor is written under `$SIMY_HOME/work-pattern-history-sync/`.
 
 The CLI starts a loopback HTTP agent on a random available port, keeps a
 48-hour, origin-scoped local session, and opens a native coding chat. The first
-chat message creates the remote audit ledger and starts the local Agentic Loop
+chat message creates the remote audit ledger and starts the local Delivery Loop
 state machine; SIMY Web mirrors the lifecycle but is not required as the task
 entry point. The CLI also accepts backend-issued launch challenges for Web-led
 runs. It builds the requirement charter, runs Codex or Claude Code, audits
 structured completion evidence, and re-instructs the executor while verification shows useful progress.
 
-Before Web creates an Agentic Loop, it classifies whether the request actually
+Before Web creates an Delivery Loop, it classifies whether the request actually
 needs a managed PR, verification, evidence, and merge lifecycle. A bounded
 ordinary task is sent to CLI 0.2.2 or newer through the one-time direct executor
 instead. CLI independently recomputes the guardrail, restricts read-only work at
 the Provider command boundary, and permits exactly one Provider invocation. It
-does not create an Agentic Loop ledger, audit session, or retry lifecycle for
+does not create an Delivery Loop ledger, audit session, or retry lifecycle for
 that task.
 
 ## Local Tasks
@@ -568,13 +570,13 @@ process running as the same OS user.
 
 
 CLI API contract v9 promotes non-coding work to a first-class **Local Task**
-runtime. A Local Task is separate from Agentic Loop: it runs one bounded Codex
+runtime. A Local Task is separate from Delivery Loop: it runs one bounded Codex
 or Claude Code session without Git branches, pull requests, audits, or automatic
 Provider retries. Knowledge-work deliverables use an isolated task sandbox by
 default, so they do not require a repository.
 
 `POST /v1/local-tasks/start` accepts JSON or the same verified multipart
-attachment envelope used by Agentic Loop and requires a stable
+attachment envelope used by Delivery Loop and requires a stable
 `Idempotency-Key`. Repeating the same request returns the original task; using
 the same key for different content returns
 `LOCAL_TASK_IDEMPOTENCY_CONFLICT`. `GET /v1/local-tasks/reconcile` with the same
@@ -664,10 +666,12 @@ original argv once. A process-wide nonce/version guard prevents update loops.
 Major releases are never installed at startup: `simy update` asks for explicit
 confirmation (`simy update --yes` is available for non-interactive automation).
 
-The macOS desktop app runs from `/Applications/SIMY.app`. Its signed updater
-keeps downloaded releases in the user cache, then replaces the desktop bundle
-after the app and its executor have stopped. The previous bundle is retained
-for recovery. If permissions or verification prevent replacement, SIMY shows
+The macOS desktop app starts from `/Applications/SIMY.app`. Its signed updater
+keeps downloaded releases in the user cache. An explicit window restart opens
+the validated immutable release while retaining its execution owner. Whole-bundle
+replacement waits for processes using the bundle to exit; a busy bundle is left
+in place and the selected release opens directly. The previous bundle is retained
+for recovery. If permissions or verification prevent installation, SIMY shows
 an installer recovery action instead of reporting the desktop app as current.
 Concurrent handoffs share one exclusive installation lock. Packages before
 0.5.29 may require one system-authorized installer update because their bundle
@@ -724,7 +728,7 @@ Auto-update E2E screenshots are generated locally under
 
 ## Executor compatibility
 
-SIMY checks the selected executor with `--version` before an Agentic Loop can
+SIMY checks the selected executor with `--version` before an Delivery Loop can
 start. The currently validated compatibility floors are Codex `0.144.0` and
 Claude Code `2.1.200`. A missing, older, or unversioned executable fails the
 preflight and returns the installed version, required version, and a recovery
@@ -846,7 +850,7 @@ permissions still require validation in the corresponding environment.
 
 Normal runs do not require a separate sync command: the CLI
 uses a verified local cache and revalidates it with TTL and ETag semantics.
-The Agentic Loop runs min-check after each code modification and the full check
+The Delivery Loop runs min-check after each code modification and the full check
 before PR readiness. Findings and the redacted signed-proof summary are recorded
 in its audit; source text is never uploaded.
 
@@ -888,7 +892,7 @@ list.
 
 ## Interactive console
 
-Running `simy` in a terminal opens the Agentic Loop chat. It discovers the
+Running `simy` in a terminal opens the Delivery Loop chat. It discovers the
 current GitHub repository and branch, selects an available executor, and keeps
 the composer active so a natural-language request can start immediately. Each
 run appears in a selectable list with its lifecycle state, process ID, and a
@@ -988,7 +992,7 @@ Publishing is handled by GitHub Actions in `.github/workflows/publish.yml`.
 
 ### Repository system scopes
 
-Version 0.5.7 supports signed `system_scope` snapshots from Quality Monitor.
+Version 0.5.7 supports signed `system_scope` snapshots from SQM.
 Each check pins the repository's connected component and graph revision. A new
 check revalidates scoped bundles online; a stale or unavailable scope cannot fall
 back to an older cached component. Minimum and full check proofs retain the
@@ -1035,13 +1039,13 @@ No twelve-view specification, reconstruction review, traceability matrix or new
 review package is required. Check actual ease, clarity and accuracy using the
 [UX guidance](src/orchestrator/delivery/references/ux-specification.md).
 
-## Purpose-driven automatic continuation
+## Purpose-driven Delivery Loop
 
 The desktop **Continue automatically / 自動で続ける** switch enables purpose-driven
 search: research, experiments, minimal implementation and direct outcome checks.
 A separate review is used for semantic inspection or when explicitly requested. Failed methods trigger replanning; review findings return to repair.
 Independent authorized work can continue while a controller-tracked operation waits.
-The existing Codex auto-continue switch enables read-only diagnosis for uncertain or
+The existing Codex **Continue automatically** switch enables read-only diagnosis for uncertain or
 information-waiting conversations, then assessed work. No default time/turn cap is
 added; explicitly enrolled purpose budgets and user stops are preserved. Native
 provider approval waits still require the original provider boundary to settle.
@@ -1082,7 +1086,7 @@ model responses or credentials. Historical summaries are not a replacement for
 signed proof files or delivery acceptance. Records from versions that kept
 analysis only in memory cannot be recovered after that process exits.
 
-The CLI Agentic Loop uses [proportionate retry and review decisions](docs/stop-prevention/proportionate-continuation.md): reasoning notes are advisory, and an independent reviewer may explicitly defer or dismiss minor/warning/info findings with a nonempty explanation while retaining them in the receipt. Required acceptance evidence and major safety findings remain protected.
+The CLI Delivery Loop uses [proportionate retry and review decisions](docs/stop-prevention/proportionate-continuation.md): reasoning notes are advisory, and an independent reviewer may explicitly defer or dismiss minor/warning/info findings with a nonempty explanation while retaining them in the receipt. Required acceptance evidence and major safety findings remain protected.
 
 ### Local coding attempt result collection
 
@@ -1398,8 +1402,8 @@ Search covers the full available catalog, including older conversations. Use
 Each row identifies its provider; identical IDs in Codex and Claude remain
 separate rows. A provider failure preserves the other provider's results with
 an explicit incomplete warning (`complete: false` and `warnings` in JSON).
-This view respects both saved provider connection settings and never enables
-automatic continuation. Disconnected providers are not read.
+This view respects both saved provider connection settings and never turns on
+Delivery Loop. Disconnected providers are not read.
 
 The terminal command reads this host and hosts explicitly registered in
 `~/.simy/remote-hosts.json`. Use `--local-only` to skip SSH, and
@@ -1461,7 +1465,7 @@ unsupported-platform error for Claude native control.
 
 Keep Claude available in an unlocked desktop while the operation runs. SIMY
 opens the exact Code-tab conversation and uses Claude's own archive/restore
-control. Running conversations and automatic continuation must be stopped first.
+control. Running conversations and Delivery Loop must be stopped first.
 Before opening that conversation, SIMY checks Claude's visible windows for
 unsent input, generating responses and dialogs. It leaves the current conversation
 in place if any are present or the window/input state cannot be inspected. The
@@ -1483,7 +1487,7 @@ with `--provider codex`.
 ### Delivery entry routes
 
 Crew owns `sessionLoopOptions` in the distributed loop kernel. Codex enrollment,
-Continue, follow-up, desktop controls and automatic continuation use purpose v2
+Continue, follow-up, desktop controls and Delivery Loop use purpose v2
 by default; old false/v1 enrollment flags cannot select the retired mode.
 Explicit strict/balanced policies and human acceptance criteria remain binding.
 Compatible saved tasks migrate without replacing identity, history or budget.
@@ -1530,7 +1534,7 @@ occupancy without accepting the task. Long-running tasks transmit current
 recovery context and recent work; the full history remains in durable state
 and can be inspected through `delivery status`.
 
-### Reusable Auto Continue evaluation inputs
+### Reusable Delivery Loop evaluation inputs
 
 For the 20 versioned conversation regression cases, immutable stopping-protocol
 plans, and pinned SWE-bench Pro public-task preparation, see

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@awak-app/simy-cli",
-  "version": "0.7.35",
+  "version": "0.7.37",
   "description": "Local SIMY executor for non-coding Local Tasks and Agentic Loop runs.",
   "type": "module",
   "bin": {

package/src/agent.js

@@ -3126,7 +3126,7 @@ export async function startAgent({
               request,
             });
         const runId = String(remoteRun?.id || "");
-        if (!runId) throw new Error("SIMY Web did not return an Agentic Loop run id.");
+        if (!runId) throw new Error("SIMY Web did not return a Delivery Loop run id.");
         const run = createRun({ runId, request, session, apiOrigin, localStateRoot: codingRunStateRoot, sqmHistoryRoot: sessionRoot });
         await saveCodingRunSnapshot(codingRunStateRoot, run.snapshot);
         registry.create(run);

package/src/backend-executable.js

@@ -28,7 +28,7 @@ export async function inspectBackendExecutable(
   backend,
   { spawnImpl = spawn, candidates, timeoutMs = 5_000, signal, discoveryOptions = {}, minimumVersion, executionEnvironment } = {},
 ) {
-  const assertActive = () => { if (signal?.aborted) throw Object.assign(new Error('Backend inspection was cancelled.'), { code: 'cancelled' }); };
+  const assertActive = () => { if (signal?.aborted) throw Object.assign(new Error('AI provider inspection was cancelled.'), { code: 'cancelled' }); };
   assertActive();
   const policy = versionPolicy(backend);
   const requiredVersion = minimumVersion || policy.minimumVersion;

package/src/cli-contract.js

@@ -14,7 +14,7 @@ import {
 } from "./durable-local-task-steps-contract.js";
 import { buildExecutionCapabilityContract } from "./execution-capability-contract.js";
 
-export const CLI_VERSION = "0.7.35";
+export const CLI_VERSION = "0.7.37";
 export const CLI_API_CONTRACT_VERSION = 10;
 
 export function narrativeImageCapabilityMetadata() {

package/src/cli-observability.js

@@ -7,7 +7,7 @@ import { execFile } from 'node:child_process';
 import { promisify } from 'node:util';
 import { webApiHeaders, webApiUrl } from './web-api.js';
 // Keep this leaf module package-safe: hooks load it before the full CLI contract.
-const CLI_VERSION = '0.7.35';
+const CLI_VERSION = '0.7.37';
 
 const STORE = 'cli-observability-state.json';
 const LEGACY_CONFIG = 'cli-observability.json';

package/src/codex-plugins/simy/.codex-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "simy",
   "version": "0.1.0",
-  "description": "SIMY Auto Continue, quality checks and PR review through the original conversation's existing controls.",
+  "description": "SIMY Delivery Loop, SQM checks and PR review through the original conversation's existing controls.",
   "author": {
     "name": "Specialist Doctors Inc.",
     "url": "https://github.com/SpecialistDoctors-Inc"
@@ -17,17 +17,17 @@
   "interface": {
     "displayName": "SIMY",
     "shortDescription": "Continue, check and review.",
-    "longDescription": "Ask SIMY to continue unfinished work in the original conversation through the running CLI app's existing Auto Continue controller, or request quality checks and PR reviews. Read the same real work state and next action as the app. Saved delegation is not execution proof. Stops, permission waits and opt-outs are preserved. Retired local delivery agents stay retired.",
+    "longDescription": "Ask SIMY to continue unfinished work in the original conversation through the running CLI app's existing Delivery Loop controller (Continue automatically), or request SQM checks and PR reviews. Read the same real work state and next action as the app. Saved delegation is not execution proof. Stops, permission waits and opt-outs are preserved. Retired local delivery agents stay retired.",
     "developerName": "Specialist Doctors Inc.",
     "category": "Developer Tools",
     "capabilities": [
-      "Auto Continue",
+      "Delivery Loop",
       "Quality",
       "Review"
     ],
     "defaultPrompt": [
-      "Enable SIMY Auto Continue for this conversation and show its actual progress.",
-      "Run the SIMY SQM Loop for the current thread.",
+      "Continue this conversation automatically with SIMY and show its actual progress.",
+      "Run a SIMY SQM check for the current thread.",
       "Review the current pull request with SIMY."
     ],
     "composerIcon": "./assets/logo.png",

package/src/codex-plugins/simy/hooks/README.md

@@ -27,7 +27,7 @@ including when SessionStart was missed. No per-project enable command is needed.
 preserve it. Existing loops, explicit stops, session opt-outs and revoked provider
 consent remain authoritative. No hook grants provider trust. Stop checks that
 session's checkpoint and returns bounded next-step instructions. Approval/input
-blocks, interruption, completion, failures and missing progress stop automatic continuation.
+blocks, interruption, completion, failures and missing progress stop Delivery Loop.
 Provider errors/exits are surfaced to SIMY's desktop; a dead provider is not
 silently restarted. No hook writes trust, permission settings, or SQM evidence.
 
@@ -35,7 +35,7 @@ Private state is under `SIMY_HOME/delivery-hooks` (default `~/.simy/delivery-hoo
 keyed by provider and session ID and bound to the canonical working directory.
 Historical hook observations do not prove hooks are still enabled. See the
 CLI delivery command documentation for checkpoint commands and explicit recovery.
-The local Codex delivery-loop skill is retired; request the CLI Agentic Loop
+The local Codex delivery-loop skill is retired; request the CLI Delivery Loop
 explicitly for new delivery work. Existing lifecycle hooks do not start it.
 
 SIMY does not limit a task's elapsed time, number of turns or provider tokens.
@@ -47,7 +47,7 @@ criteria cannot manufacture progress, and activity alone does not prove it.
 Required approvals, manual stops and unreconciled effects still block dispatch.
 
 Context-first tasks without host-checkable acceptance pause rather than fail at
-that limit, so Auto-continue stays available after review. An explicit resume,
+that limit, so Delivery Loop stays available after review. An explicit resume,
 answer or approval resets both no-progress attempt counters and the progress clock,
 even before a threshold was reached. Failure counts and verified criterion history
 remain; reusing old evidence cannot count as fresh progress.

package/src/codex-plugins/simy/hooks/delivery-loop.mjs

@@ -996,7 +996,7 @@ function hookDeliveryContext(record, automatic = false) {
   const identity = `SIMY connected to this ${provider} conversation: ${session}.`;
   if (!run && automatic && !record.automaticStartBlocked && !record.providerArchived && !record.parentSession && !(record.autoContinue === false && record.autoContinueEpoch)) return `${identity} Automatic Delivery Loop start is enabled for this workspace. The next nonempty user prompt starts checkpoint tracking in this conversation; do not issue a duplicate start. Provider permissions and user stops remain authoritative.`;
   if (!run) return `${identity} Delivery Loop has not started. When the user requests it, use simy delivery start --provider ${provider} --session ${session} with the agreed objective and next step. For future conversations in this workspace, the user can enable automatic start once with simy delivery auto on --provider ${provider}. Never enable, trust, resume, or reset a loop without the user's request.`;
-  return `${identity} Existing loop: ${run.state}. ${deliveryAcceptanceContext(run)} ` +
+  return `${identity} Existing Delivery Loop: ${run.state}. ${deliveryAcceptanceContext(run)} ` +
     (['active', 'reported_done', 'completed'].includes(run.state) ? `${correctiveFollowupInstruction} ` : '') +
     (provider === 'codex' && ['completed', 'reported_done'].includes(run.state) ? `Finished run ID: ${run.id}. Its completion does not supervise later user requests. If the user explicitly requests supervision of the current goal, use simy delivery followup --provider codex --session ${session} --previous-run-id ${run.id} --objective CURRENT_GOAL. A different goal archives the old run; the same ordinary reported_done goal retains identity and budgets for read-only reconciliation. A new prompt alone is not consent to run this command. ` : '') +
     (run.state === 'active' && !run.managed ? `Before ending the turn, submit an honest JSON checkpoint with simy delivery checkpoint --provider ${provider} --session ${session} --file -. Use status in_progress with summary, remainingWork and nextStep for remaining work; completed with summary, acceptanceMet:true and evidence for finished work; or blocked with summary and blocker missing_information or authorization. The first completed checkpoint requests an outcome check in this conversation; inspect the actual result and submit a fresh checkpoint. If requested work remains, perform the first authorized step instead of ending with self-evaluation or a plan. Completion is provider-reported, not independently verified. ${informationWaitInstruction} ` : '') +
@@ -1051,7 +1051,7 @@ export async function startDelivery({ root = rootDirectory(), provider, session,
     if ((following || reconciling) && (provider !== 'codex' || !admitted || !managedState?.autoRecovery || managedState.capability !== 'codex_app_server' || managedState.initialAction.kind !== 'read' || managedState.contextBootstrap)) throw new Error('Follow-up requires exact-session managed supervision, a concrete goal and read-only reconciliation.');
     if (upgrading && (!managedState?.autonomyPolicy || managedState.initialAction.kind !== 'read' || managedState.contextBootstrap)) throw new Error('Upgrade requires read-only reconciliation and frozen acceptance of the original objective.');
     if (!managedState && continuationMode(continuationPolicy) === 'progress') throw new Error('Progress continuation requires frozen managed acceptance criteria.');
-    if (followupAutoContinue && !(following || reconciling)) throw new Error('Auto-continue follow-up requires the exact finished run.');
+    if (followupAutoContinue && !(following || reconciling)) throw new Error('Delivery Loop follow-up requires the exact finished run.');
     if (autoContinueEpoch && managedState) managedState.autoContinueEpoch = autoContinueEpoch;
     if (managedState && previous.associationTaskId) {
       managedState.taskId = previous.associationTaskId;

package/src/codex-plugins/simy/skills/continue/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: continue
-description: Start SIMY Auto Continue for the original conversation when the user explicitly asks SIMY or @simy to take over or continue unfinished work. Also read its real execution status or turn automatic continuation off. This is a controller entry, not an independent delivery agent or an SQM check.
+description: Turn on SIMY Delivery Loop (Continue automatically) for the original conversation when the user explicitly asks SIMY or @simy to take over or continue unfinished work. Also read its real execution status or turn Delivery Loop off for this conversation. This is a controller entry, not an independent delivery agent or an SQM check.
 ---
 
 # Continue the original conversation with SIMY
@@ -32,8 +32,8 @@ existing conversation reference before changing anything.
 Use this command on the original local host. Preserve a remote selection's host
 and use its existing SIMY remote controls instead of treating it as local.
 
-This command delegates to the running SIMY app's existing Auto Continue control
-and background coordinator. It preserves the original goal, task history,
+This command delegates to the running SIMY app's existing Delivery Loop control
+(the Continue automatically switch) and background coordinator. It preserves the original goal, task history,
 corrections, permissions, failure budgets and approvals. It does not start a
 second provider process, create a replacement task, change accounts or models,
 trust hooks, or grant native permissions. The original current turn continues
@@ -51,7 +51,7 @@ result and acceptance. Continue already-authorized work in this conversation.
 To inspect again, use the same exact target with `auto-continue status` in place
 of `auto-continue on`. If the response was lost or status reading failed after
 saving, query status; never automatically repeat a mutation to infer success.
-When asked to turn automatic continuation off, use `auto-continue off` for the
+When asked to turn Delivery Loop off, use `auto-continue off` for the
 same target. OFF preserves the current native turn; an explicit stop uses the
 existing task Stop control. A stopped, rejected, archived, settled, disconnected
 or unavailable task stays within its existing boundary: report the returned

package/src/codex-plugins/simy/skills/continue/agents/openai.yaml

@@ -1,6 +1,6 @@
 interface:
-  display_name: "SIMY Auto Continue"
-  short_description: "Continue the original task through SIMY"
-  default_prompt: "Use $simy:continue to enable SIMY Auto Continue for this conversation and report its actual work state."
+  display_name: "SIMY Continue automatically"
+  short_description: "Turn on Delivery Loop for this conversation"
+  default_prompt: "Use $simy:continue to turn on SIMY Delivery Loop for this conversation and report its actual work state."
 policy:
   allow_implicit_invocation: true

package/src/codex-plugins/simy/skills/diagnose/SKILL.md

@@ -1,9 +1,9 @@
 ---
 name: diagnose
-description: Diagnose a stopped or unresponsive SIMY task when the user says “SIMY CLIに相談して”, asks why it stopped, or requests manual SIMY diagnosis. Distinguish read-only diagnosis from an explicit request to repair and continue the original work.
+description: Check why it stopped (停止理由の確認) for a stopped or unresponsive SIMY task when the user says “SIMY CLIに相談して” or asks why it stopped. Distinguish a read-only check from an explicit request to repair and continue the original work. Environment diagnosis is `simy doctor`, not this entry.
 ---
 
-# Consult SIMY CLI
+# Check why it stopped (停止理由の確認)
 
 Use this entry for an explicit consultation, diagnosis, or repair-and-continue request, independently of Delivery Loop enrollment or permission to resume. Prefer `$SIMY_HOME/bin/simy` or `$HOME/.simy/bin/simy`. Inspect `--version` and `delivery --help` to confirm `consult` is available. If only `diagnose` exists, use the legacy command below for diagnosis and report repair as unavailable; never reinterpret a failed consult invocation as execution. An older installed CLI is a compatibility gap, never permission to switch its live runtime.
 

package/src/codex-plugins/simy/skills/sqm-loop/SKILL.md

@@ -1,9 +1,9 @@
 ---
 name: sqm-loop
-description: Run the SIMY Software Quality Monitor against the repository and changes discussed in the current Codex thread. Use when the user asks SIMY to start SQM, quality-check the current work, or prepare a PR quality proof.
+description: Run a SIMY SQM check on the repository and changes discussed in the current Codex thread, on request. Use when the user asks SIMY to start SQM or quality-check the current work now. The opted-in pre-PR gate is the separate simy-sqm-pr-check skill.
 ---
 
-# SIMY SQM Loop
+# SIMY SQM check
 
 Treat the current Codex thread, its attachments, and the active repository as the request context. Convert that context into explicit quality expectations, then verify the actual repository state; do not merely summarize the conversation.
 
@@ -22,7 +22,7 @@ Investigate and repair authorized reversible setup; preserve real account,
 organization, device and provider authority. Do not guess credentials or invent a
 successful scope. A connected scope is not quality acceptance.
 
-## Run the quality loop
+## Run the SQM check
 
 1. Identify the repository root and the user-visible behavior or incident discussed in the current thread.
 2. Preserve unrelated user changes and follow repository instructions.

package/src/codex-plugins/simy/skills/sqm-loop/agents/openai.yaml

@@ -1,6 +1,6 @@
 interface:
-  display_name: "SIMY SQM Loop"
-  short_description: "Run SQM from the current Codex thread"
+  display_name: "SIMY SQM check"
+  short_description: "Run an SQM check on the current work"
   default_prompt: "Use $sqm-loop to quality-check the work in this thread."
 policy:
   allow_implicit_invocation: true

package/src/console/app.js

@@ -176,7 +176,7 @@ export function AgenticLoopConsole({ agent, onQuit = () => {}, remoteConnections
     }
     if (!selectedRun) {
       if (command.type !== "guidance") {
-        setNotice("That command requires a selected Agentic Loop run.");
+        setNotice("That command requires a selected Delivery Loop run.");
         return;
       }
       setBusy(true);
@@ -213,12 +213,12 @@ export function AgenticLoopConsole({ agent, onQuit = () => {}, remoteConnections
             setNotice("Guidance queued for the next executor handoff.");
           } else {
             await agent.controls.continue(selectedRun, command.message);
-            setNotice("Human guidance accepted. The agentic loop is continuing.");
+            setNotice("Human guidance accepted. The Delivery Loop is continuing.");
           }
           break;
         case "decision":
           await agent.controls.applyDecision(selectedRun, command);
-          setNotice("Human decision recorded. The agentic loop is continuing.");
+          setNotice("Human decision recorded. The Delivery Loop is continuing.");
           break;
         case "recheck":
           await agent.controls.recheck(selectedRun);
@@ -297,7 +297,7 @@ export function AgenticLoopConsole({ agent, onQuit = () => {}, remoteConnections
         await agent.controls.selectRepository(selectedRun, match);
         setDetailOpen(true);
         setDetailScrollFromEnd(0);
-        setNotice(`Authorized ${match.repository}. The same Agentic Loop run is continuing.`);
+        setNotice(`Authorized ${match.repository}. The same Delivery Loop run is continuing.`);
       } else if (selectedRun) {
         setNotice(
           `${selectedRun.request.repository} was not found under ${result.root}. Run /scan <path> for another authorized root.`,
@@ -328,7 +328,7 @@ export function AgenticLoopConsole({ agent, onQuit = () => {}, remoteConnections
         setRepositoryPicker(null);
         setDetailOpen(true);
         setDetailScrollFromEnd(0);
-        setNotice(`Authorized ${selected.repository}. The same Agentic Loop run is continuing.`);
+        setNotice(`Authorized ${selected.repository}. The same Delivery Loop run is continuing.`);
       } else {
         const verified = await agent.controls.selectRepository(null, selected);
         setDraft((current) => ({
@@ -510,7 +510,7 @@ export function AgenticLoopConsole({ agent, onQuit = () => {}, remoteConnections
       if (confirmStop) void execute("/stop");
       else {
         setConfirmStop(true);
-        setNotice("Press x again to stop the selected Agentic Loop run.");
+        setNotice("Press x again to stop the selected Delivery Loop run.");
       }
       return;
     }
@@ -641,7 +641,7 @@ function NewTask({ draft, loginUrl, width }) {
         dimColor: true,
         wrap: "wrap",
         children:
-          "Describe the outcome. SIMY will choose a one-time Local Task or an Agentic Loop.",
+          "Describe the outcome. SIMY will choose a one-time Local Task or a Delivery Loop.",
       }),
       jsx(Text, { children: " " }),
       jsx(Text, {