sneakoscope 6.1.0 → 6.2.0

package/README.md

@@ -13,7 +13,7 @@ sneakoscope is a proof-first Codex trust layer for bounded agent workflows, sear
 
 # Sneakoscope Codex
 
-**The proof-first swarm harness for Codex. Machine-verified completion, not vibes.**
+**The proof-first workflow harness for Codex. Machine-verified completion, not vibes.**
 
 [![npm version](https://img.shields.io/npm/v/sneakoscope?color=cb3837&logo=npm)](https://www.npmjs.com/package/sneakoscope)
 [![node](https://img.shields.io/badge/node-%3E%3D20.11-339933?logo=node.js&logoColor=white)](#requirements)
@@ -23,9 +23,9 @@ sneakoscope is a proof-first Codex trust layer for bounded agent workflows, sear
 
 </div>
 
-Sneakoscope Codex (`sks`) is a Codex CLI and Codex App harness for people who want parallel AI coding without losing proof. It gives Codex a simple front door, a dynamic worker swarm, a local dashboard, TriWiki project memory, and release gates that separate machine evidence from LLM opinion.
+Sneakoscope Codex (`sks`) is a Codex CLI and Codex App harness for people who want parallel AI coding without losing proof. It gives Codex a simple front door, an official subagent workflow, a local dashboard, TriWiki project memory, and release gates that separate machine evidence from LLM opinion.
 
-Current release: SKS **6.1.0**. Normal Codex sessions continue to inherit the complete Codex catalog without silent replacement. Naruto workers now use a scoped, live-catalog-verified GPT-5.6 policy: Terra `xhigh/max` for coding, Sol `max` for refactoring/planning/strategy/integration, and Luna `xhigh/max` for E2E/browser/Computer Use/GUI verification. This release restores the Codex Desktop Fast selector under codex-lb; its `1.5x` multiplier follows the [official Codex Speed documentation](https://learn.chatgpt.com/docs/agent-configuration/speed), not an SKS benchmark. It also self-repairs native capability plugins and hardens Voxel TriWiki retention: command help stays read-only, closed routes are not kept active, recent sessions receive a two-hour grace window, mission compaction preserves durable evidence byte-for-byte and deletes only known disposable runtime files, legacy gzip archives are transparently SHA-256-verified when read, full state-budget scans cover the complete top-level state surface, and release-gate history retains five runs by default. See [CHANGELOG.md](CHANGELOG.md).
+Current release: SKS **6.2.0**. `$Naruto` is the SKS alias for the Codex official subagent workflow. The parent runs on GPT-5.6 Sol Max. Child routing is fixed to Luna Max for tiny short-context mechanical work, Sol High for ordinary UI/logic/backend/native implementation, Sol Max for review/debugging/planning/security and other judgment, and Terra Medium for long-context analysis plus Computer Use, Browser/Chrome, and image-generation execution. Official `SubagentStart`/`SubagentStop` events prove lifecycle only; successful completion additionally requires a structured parent summary with an explicit outcome for every thread. Preparation context alone is never completion proof. Normal non-Naruto Codex sessions continue to preserve explicit user model choices. See [CHANGELOG.md](CHANGELOG.md).
 
 ## Install
 
@@ -42,19 +42,39 @@ npm i -g sneakoscope
 sks doctor --fix
 ```
 
+The SKS menu bar shows the installed Codex CLI version and latest known version. An `⬆` marker appears when an update is available; **Update Codex CLI Now** uses native `codex update` when the selected CLI advertises it, otherwise it verifies the installation provenance and invokes the matching official standalone-installer, npm-global, or Homebrew-cask update method. If the method cannot be verified, it fails closed instead of guessing. This is an explicit global tool mutation. **Run sks doctor --fix** performs the global-only menu repair flow without treating the user's home directory as a project.
+
+**Manage MCP Servers…** opens a native macOS manager for the global `~/.codex/config.toml`. It can add remote URL or local stdio servers, enable/disable existing entries, remove entries after confirmation, and refresh the current state. Mutations are lock-protected, backed up, TOML-validated, and written with mode `0600`; configured environment values and command arguments are never rendered in the list. Changes apply to new Codex sessions. The same plumbing is available through `sks menubar mcp list|add|enable|disable|remove` for diagnostics and tests.
+
 ## The Front Door
 
 | Command | What it does |
 | --- | --- |
 | `$Plan "task"` | Planning only. Writes `.sneakoscope/plans/<slug>.md`; no code edits. |
-| `$Work` | Executes the latest plan through evidence-gated SKS work. |
-| `$Swarm "task"` | Runs the Naruto dynamic parallel swarm with machine verification. |
+| Explicit `$Work` | Executes the latest plan through evidence-gated SKS work. Ordinary prose containing “work” is not treated as this alias. |
+| `$Swarm "task"` | Compatibility alias for the Naruto Codex official subagent workflow. |
 | `$Team "task"` | Deprecated v5 compatibility alias. New execution redirects to `$Naruto`; legacy Team observe/watch commands remain available for old missions. |
 | `$MAD-SKS` / `sks mad-sks` | Single high-risk MAD route for scoped permission widening plus SQL-plane execution, including read-back proof and profile closure. |
 | `$MAD-DB` / `sks mad-db` | Deprecated compatibility alias. Translates to `$MAD-SKS` sql-plane commands for one release. |
 | `$Review` / `sks review --staged` | Reviews diffs with `evidence: machine` findings sorted above `evidence: llm`. |
 | `sks ui` | Opens the local live dashboard at `http://127.0.0.1:4477`. |
 
+`sks --mad` now prioritizes the interactive ready path: independent macOS config probes run concurrently, failed read-only preflight does not repeat mutation-capable repair inspection, verified Zellij/codex-lb evidence is reused, and the remote Zellij update lookup runs after the UI is ready. Existing unreadable or malformed config still blocks safely; pass an explicit repair flag such as `--repair-config` when repair is intended.
+
+## Naruto In 6.2.0
+
+`$Naruto` and `sks naruto run "task" --agents 8 --max-threads 12` use Codex official subagents. The parent is GPT-5.6 Sol Max. Tiny mechanical `worker` slices use Luna Max; ordinary UI, logic, backend, and native coding uses Sol High; review, testing, debugging, architecture, integration, security, database, research, release, and other judgment-sensitive work uses Sol Max; long-context scans and direct Computer Use, Browser/Chrome, or image-generation execution uses Terra Medium. Mixed execution/judgment work is split when possible, and unsplittable judgment defaults to Sol Max.
+
+Fresh SKS-owned project config uses `max_threads = 12`, `max_depth = 1`, `job_max_runtime_seconds = 1200`, and `interrupt_message = true`. `max_depth = 1` means subagents cannot spawn nested subagents. Explicit user limits are preserved, and larger requests run in waves.
+
+Gates are task-profile aware: greetings and answer-only turns create no mission gate; tiny work gets minimal verification; parallel work gets scoped ownership and verification; high-risk work keeps the full safety gates. `SubagentStart`/`SubagentStop` prove lifecycle only. Completion also requires `subagent-parent-summary.json` with one trustworthy structured outcome per thread, correlated with `subagent-events.jsonl` and `subagent-evidence.json`.
+
+Every installed Codex hook runs one common Naruto decision gate. The gate records `none`, `generic_naruto`, or `route_owned`: Answer, DFix, Wiki, Computer Use, Goal, and simple Git/control turns stay lightweight; ordinary non-trivial work defaults to two independent official subagents; critical work spanning at least three risk domains may use three. Research, AutoResearch, and QA-Loop retain their own exact orchestration contracts instead of receiving a second generic fan-out. Explicit `--agents N` remains authoritative.
+
+SKS installs twenty-five narrow project custom agents, including native AppKit, toolchain, protocol, runtime-reliability, TriWiki-evidence, long-context, Computer Use, Browser/Chrome, and image-generation specialists. Delegation prompts inject at most the three roles recommended for the current goal rather than serializing the full catalog, so expanding role coverage does not serialize the full inventory into every prompt. TriWiki context is also bounded and query-aware: ordinary work receives up to four trust/hydration anchors and complex, parallel, or high-risk work receives up to six, with source hydration required before relying on lower-trust hints. In CLI Zellij mode, the right side is a live observability surface rather than a static lane reservation: one monitor plus one viewport by default (maximum three) shows official thread role/model, redacted live phase/task/file updates from exact-agent Codex 0.144.1+ rollouts, `running`, `verifying`, and trustworthy parent-verdict completion/failure states. Rollout activity is display-only and never completion proof.
+
+`--clones` and `workers` remain one-release spelling aliases for official subagent behavior. The historical Naruto process swarm and its environment opt-in are removed; legacy backend, scheduler, pool, and model flags fail closed.
+
 ## Why Not Just An LLM Reviewer?
 
 | Question | Oracle-style LLM review | SKS gate/review |
@@ -62,14 +82,14 @@ sks doctor --fix
 | Did tests/typecheck fail? | Another model may say so. | Machine check output is tagged `evidence: machine`. |
 | Are findings ranked? | Usually one blended opinion. | Machine evidence sorts before LLM findings. |
 | Can work stop? | The model decides. | Stop gates, Completion Proof, and Honest Mode decide. |
-| Can I watch the swarm? | Usually no runtime UI. | Zellij panes plus `sks ui` SSE dashboard. |
+| Can I inspect agent-thread progress? | Usually no runtime UI. | `sks ui` plus the official Codex subagent/thread surfaces. |
 
 ## Dashboard
 
 `sks ui` serves a dependency-free localhost dashboard with:
 
 - mission, route, elapsed time, and gate badge
-- live worker slot grid with role, backend/model badge, progress, and current task
+- live official subagent thread grid with role/model badge, progress, and current task
 - run/verify/queue/done/fail counters
 - recent mission events
 - current gate checklist
@@ -84,19 +104,20 @@ The reproducible VHS script lives at [docs/demo.tape](docs/demo.tape).
 vhs docs/demo.tape
 ```
 
-It shows the v5 flow: one-line install, `$Plan`, `$Work`/`$Swarm`, `sks review`, and `sks ui`.
+It shows the current quickstart flow: one-line install, `$Plan`, `sks review`, `sks ui`, and an official `$Naruto` subagent run.
 
 ## Proof Surfaces
 
-- Dynamic swarm: `sks naruto run "task" --clones 14 --json`
+- Official subagents: `sks naruto run "task" --agents 14 --max-threads 12 --json`
 - Review report: `.sneakoscope/reports/review-report.json`
 - Harness benchmark: `.sneakoscope/reports/harness-benchmark.json`
 - Project memory: `sks memory build`
 - Codebase index/pack for LLM context: `sks wiki refresh --code`, `sks wiki validate --json` (code-pack freshness)
 - Native capability repair: `sks doctor --fix` (imagegen/Computer Use/Browser Use), `.sneakoscope/reports/native-capability-readiness.json`
+- codex-lb continuity: `sks codex-lb status --json` verifies the selected proxy's unauthenticated `/health` `X-App-Version`. Tool-heavy continuation requires codex-lb `1.21.0-beta.3` or later; older or unverified deployments block setup, doctor, and launch instead of silently falling back.
 - Agent bridge for any agent system: `sks mcp-server`, `sks agent-bridge setup`, `SKS_AGENT_MODE=1` — see [docs/AGENT-BRIDGE.md](docs/AGENT-BRIDGE.md)
-- Release gates: `node ./dist/scripts/release-gate-dag-runner.js --preset release --full`
-- Lifecycle-disabled npm release: run `npm run release:check:full`, then `npm run publish:prep-ignore-scripts`, and verify with `npm publish --dry-run --ignore-scripts --json` before the intentional `npm run publish:ignore-scripts` publication wrapper.
+- Release gates: `npm run release:check:affected` for ordinary change-aware verification and `npm run release:check:confidence` for the final local confidence pass.
+- Release preparation handoff: run `npm run typecheck --silent`, one `npm run build:clean --silent`, focused tests for the changed surfaces, `npm run release:check:affected`, `npm run release:check:confidence`, and finally `npm pack --dry-run --ignore-scripts --json`. Publish authorization and publication remain a distinct repository-maintainer workflow.
 - Release readiness notes: [docs/release-readiness.md](docs/release-readiness.md) and [CHANGELOG.md](CHANGELOG.md)
 - Image generation review routes require Codex App `$imagegen`/`gpt-image-2` evidence with recorded output hashes; direct API fallback and mock fixtures do not satisfy full route gates.
 
@@ -110,7 +131,10 @@ It shows the v5 flow: one-line install, `$Plan`, `$Work`/`$Swarm`, `sks review`,
   - Auth/provider changes wait for the ChatGPT/Codex app to exit and reopen by bundle id; a failed restart is reported as a failed menu action.
   - Update installs always rebuild the companion with the newly installed SKS package, preventing a previous-version updater from restoring a stale menu binary.
   - The menubar dropdown's `View Last Log` item opens the most recent background action's log file, so you don't need to keep a Terminal window open to see command output.
+  - `Manage MCP Servers…` provides a resizable native table and add/remove/enable/disable controls for global Codex MCP configuration. Secret environment values and command arguments are accepted through native dialogs/stdin but omitted from list output and logs.
   - `sks menubar status --json` reports a `codex_sync` object with `bundle_id`, `codex_running`, and `icon_visible_expected` to show Codex-lifecycle detection state.
+  - The menu displays the installed Codex CLI version, adds an `⬆` status icon when `sks codex update-status` sees a newer release, runs the official self-updater through `Update Codex CLI Now`, and exposes `Run sks doctor --fix` as a background repair action.
+- If Codex shows `[No tool output found for custom tool call ...]`, SKS blocks reuse of that structurally ambiguous thread. Upgrade codex-lb (or explicitly run `sks codex-lb use-oauth`), inspect possible side effects, then continue the persisted mission in a fresh Codex task. SKS never rewrites session JSONL or fabricates a successful tool output.
 - Zellij optional but recommended for terminal worker panes
 
 ## License

package/crates/sks-core/Cargo.lock

@@ -76,7 +76,7 @@ dependencies = [
 
 [[package]]
 name = "sks-core"
-version = "6.1.0"
+version = "6.2.0"
 dependencies = [
  "serde_json",
 ]

package/crates/sks-core/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "sks-core"
-version = "6.1.0"
+version = "6.2.0"
 edition = "2021"
 
 [dependencies]

package/dist/bin/fast-inline.js

@@ -32,6 +32,21 @@ function doctorJsonFastInline() {
             hook_evidence_policy: 'unknown-do-not-count',
             agent_role_strategy: 'message-role'
         },
+        zellij_readiness: {
+            schema: 'sks.zellij-readiness.v1',
+            binary: 'zellij',
+            status: 'skipped',
+            min_version: '0.41.0',
+            version: null,
+            required_for: ['sks --mad', 'interactive lane UI'],
+            layout_proof: 'unavailable',
+            pane_proof: 'unavailable',
+            screen_proof: 'unavailable',
+            tmux_removed_runtime: true,
+            mad_ready: false,
+            cli_ready: false,
+            ready_for_interactive_runtime: false
+        },
         codex: { bin: null, version: null, available: null, skipped: true, reason: 'fast_readonly_json' },
         repair: {
             setup: null,
@@ -42,15 +57,9 @@ function doctorJsonFastInline() {
         warnings: ['fast_readonly_doctor_skipped_optional_deep_diagnostics']
     }, null, 2)}\n`);
 }
-function narutoHelpJsonFastInline() {
-    process.stdout.write(`${JSON.stringify({
-        schema: 'sks.naruto-command-result.v1',
-        ok: true,
-        action: 'help',
-        fast_path: true,
-        usage: 'sks naruto run "<prompt>" [--json] [--mock] [--readonly] [--clones N]',
-        commands: ['help', 'status', 'dashboard', 'workers', 'proof', 'run']
-    }, null, 2)}\n`);
+async function narutoHelpJsonFastInline() {
+    const { buildNarutoHelpResult } = await import('../core/subagents/naruto-help-contract.js');
+    process.stdout.write(`${JSON.stringify(buildNarutoHelpResult(), null, 2)}\n`);
 }
 async function hookUserPromptSubmitPerfInline() {
     const raw = await readStdinInline();

package/dist/bin/sks-dispatch.js

@@ -39,9 +39,9 @@ async function runSks(args) {
         const { hookDaemonInline } = await import('../core/daemon/sksd-hook-dispatch.js');
         await hookDaemonInline(args[1]);
     }
-    else if (args[0] === 'naruto' && (args[1] === 'help' || args.includes('--help')) && args.includes('--json')) {
+    else if (args.length === 3 && args[0] === 'naruto' && args[1] === 'help' && args[2] === '--json') {
         const { narutoHelpJsonFastInline } = await import('./fast-inline.js');
-        narutoHelpJsonFastInline();
+        await narutoHelpJsonFastInline();
     }
     else if (args[0] === 'help' || args[0] === '--help' || args[0] === '-h') {
         if (args.length > 1) {

package/dist/cli/args.js

@@ -29,6 +29,8 @@ export function positionalArgs(args = []) {
         '--project-ref',
         '--agent',
         '--agents',
+        '--clones',
+        '--max-threads',
         '--target-active-slots',
         '--work-items',
         '--minimum-work-items',

package/dist/cli/command-manifest-lite.js

@@ -20,7 +20,7 @@ export const COMMAND_MANIFEST_LITE = [
     { name: 'wizard', summary: 'Open setup wizard help', maturity: 'stable' },
     { name: 'usage', summary: 'Show focused usage topic', maturity: 'stable', readonly: true, allowedDuringActiveRoute: true, diagnostic: true },
     { name: 'quickstart', summary: 'Show quickstart flow', maturity: 'stable' },
-    { name: 'setup', summary: 'Initialize SKS state', maturity: 'stable' },
+    { name: 'setup', summary: 'Initialize SKS state', maturity: 'stable', skipMigrationGate: true },
     { name: 'bootstrap', summary: 'Initialize SKS project files', maturity: 'stable' },
     { name: 'init', summary: 'Initialize local control surface', maturity: 'stable' },
     { name: 'deps', summary: 'Check local dependencies', maturity: 'stable' },
@@ -55,7 +55,7 @@ export const COMMAND_MANIFEST_LITE = [
     { name: 'team', summary: 'Deprecated alias. New execution redirects to Naruto; legacy observe/watch remains.', maturity: 'beta', mutatesRouteState: true, deprecated: true },
     { name: 'agent', summary: 'Run native multi-session agent missions', maturity: 'beta', mutatesRouteState: true },
     { name: 'with-local-llm', summary: 'Enable or inspect local Ollama worker backend', maturity: 'beta' },
-    { name: 'naruto', summary: 'Run $Naruto shadow-clone swarm (up to 100 parallel sessions)', maturity: 'labs', mutatesRouteState: true },
+    { name: 'naruto', summary: 'Run the $Naruto Codex official subagent workflow', maturity: 'labs', mutatesRouteState: true },
     { name: 'stop-gate', summary: 'Check canonical stop-gate resolution for a route/mission', maturity: 'beta', readonly: true, skipMigrationGate: true, allowedDuringActiveRoute: true, diagnostic: true },
     { name: 'route', summary: 'Inspect or close active route state', maturity: 'beta', skipMigrationGate: true, allowedDuringActiveRoute: true, diagnostic: true },
     { name: 'loop', summary: 'Dynamic Loop Runtime: plan/run/status/proof loop graphs.', maturity: 'labs', mutatesRouteState: true },
@@ -90,7 +90,6 @@ export const COMMAND_MANIFEST_LITE = [
     { name: 'code-structure', summary: 'Scan source structure', maturity: 'labs' },
     { name: 'rust', summary: 'Inspect optional Rust accelerator status and smoke parity', maturity: 'beta' },
     { name: 'gx', summary: 'Render/validate GX cartridges', maturity: 'labs' },
-    { name: 'db', summary: 'Inspect DB safety policy', maturity: 'beta' },
     { name: 'eval', summary: 'Run eval reports', maturity: 'labs' },
     { name: 'harness', summary: 'Run harness fixtures', maturity: 'labs' },
     { name: 'wiki', summary: 'Manage TriWiki and image voxel ledgers', maturity: 'beta', skipMigrationGate: true, allowedDuringActiveRoute: true, diagnostic: true },

package/dist/cli/command-registry.js

@@ -115,7 +115,7 @@ export const COMMANDS = {
     wizard: entry('stable', 'Open setup wizard help', 'dist/core/commands/basic-cli.js', basicNoArgs('quickstartCommand')),
     usage: readOnly(entry('stable', 'Show focused usage topic', 'dist/core/commands/basic-cli.js', basicArgs('usageCommand'))),
     quickstart: entry('stable', 'Show quickstart flow', 'dist/core/commands/basic-cli.js', basicNoArgs('quickstartCommand')),
-    setup: entry('stable', 'Initialize SKS state', 'dist/core/commands/basic-cli.js', basicArgs('setupCommand')),
+    setup: skipMigrationGate(entry('stable', 'Initialize SKS state', 'dist/core/commands/basic-cli.js', basicArgs('setupCommand'))),
     bootstrap: entry('stable', 'Initialize SKS project files', 'dist/core/commands/basic-cli.js', basicArgs('bootstrapCommand')),
     init: entry('stable', 'Initialize local control surface', 'dist/core/commands/basic-cli.js', basicArgs('initCommand')),
     deps: entry('stable', 'Check local dependencies', 'dist/core/commands/basic-cli.js', subcommand(() => import(basicModule), 'depsCommand', 'dist/core/commands/basic-cli.js', 'check')),
@@ -150,7 +150,7 @@ export const COMMANDS = {
     team: routeStateMutator(entry('beta', 'Deprecated alias. New execution redirects to Naruto; legacy observe/watch remains.', 'dist/core/commands/team-command.js', argsCommand(() => import('../core/commands/team-command.js'), 'team', 'dist/core/commands/team-command.js')), ['team-gate.json', 'naruto-gate.json']),
     agent: routeStateMutator(entry('beta', 'Run native multi-session agent missions', 'dist/core/commands/agent-command.js', argsCommand(() => import('../core/commands/agent-command.js'), 'agentCommand', 'dist/core/commands/agent-command.js')), ['agent-gate.json']),
     'with-local-llm': entry('beta', 'Enable or inspect local Ollama worker backend', 'dist/core/commands/local-model-command.js', argsCommand(() => import('../core/commands/local-model-command.js'), 'localModelCommand', 'dist/core/commands/local-model-command.js')),
-    naruto: routeStateMutator(entry('labs', 'Run $Naruto shadow-clone swarm (up to 100 parallel sessions)', 'dist/core/commands/naruto-command.js', argsCommand(() => import('../core/commands/naruto-command.js'), 'narutoCommand', 'dist/core/commands/naruto-command.js')), ['naruto-gate.json', 'stop-gate.json']),
+    naruto: routeStateMutator(entry('labs', 'Run the $Naruto Codex official subagent workflow', 'dist/core/commands/naruto-command.js', argsCommand(() => import('../core/commands/naruto-command.js'), 'narutoCommand', 'dist/core/commands/naruto-command.js')), ['naruto-gate.json', 'stop-gate.json']),
     'stop-gate': readOnly(entry('beta', 'Check canonical stop-gate resolution for a route/mission', 'dist/core/commands/stop-gate-command.js', commandArgsCommand(() => import('../core/commands/stop-gate-command.js'), 'stopGateCommand', 'dist/core/commands/stop-gate-command.js'))),
     route: activeRouteDiagnostic(entry('beta', 'Inspect or close active route state', 'dist/core/commands/route-command.js', subcommand(() => import('../core/commands/route-command.js'), 'routeCommand', 'dist/core/commands/route-command.js', 'status'))),
     loop: routeStateMutator(entry('labs', 'Dynamic Loop Runtime: plan/run/status/proof loop graphs.', 'dist/core/commands/loop-command.js', subcommand(() => import('../core/commands/loop-command.js'), 'loopCommand', 'dist/core/commands/loop-command.js', 'help')), ['loop-graph-proof.json']),
@@ -185,7 +185,6 @@ export const COMMANDS = {
     'code-structure': entry('labs', 'Scan source structure', 'dist/core/commands/code-structure-command.js', subcommand(() => import('../core/commands/code-structure-command.js'), 'codeStructureCommand', 'dist/core/commands/code-structure-command.js', 'scan')),
     rust: entry('beta', 'Inspect optional Rust accelerator status and smoke parity', 'dist/commands/rust.js', directCommand(() => import('../commands/rust.js'), 'dist/commands/rust.js')),
     gx: entry('labs', 'Render/validate GX cartridges', 'dist/core/commands/gx-command.js', subcommand(() => import('../core/commands/gx-command.js'), 'gxCommand', 'dist/core/commands/gx-command.js', 'validate')),
-    db: entry('beta', 'Inspect DB safety policy', 'dist/core/commands/db-command.js', subcommand(() => import('../core/commands/db-command.js'), 'dbCommand', 'dist/core/commands/db-command.js', 'policy')),
     eval: entry('labs', 'Run eval reports', 'dist/core/commands/eval-command.js', subcommand(() => import('../core/commands/eval-command.js'), 'evalCommand', 'dist/core/commands/eval-command.js', 'run')),
     harness: entry('labs', 'Run harness fixtures', 'dist/core/commands/harness-command.js', subcommand(() => import('../core/commands/harness-command.js'), 'harnessCommand', 'dist/core/commands/harness-command.js', 'fixture')),
     wiki: activeRouteDiagnostic(entry('beta', 'Manage TriWiki and image voxel ledgers', 'dist/commands/wiki.js', directCommand(() => import('../commands/wiki.js'), 'dist/commands/wiki.js'))),
@@ -196,7 +195,7 @@ export const COMMANDS = {
     'all-features': entry('beta', 'Run all-features selftest', 'dist/commands/all-features.js', directCommand(() => import('../commands/all-features.js'), 'dist/commands/all-features.js')),
     perf: entry('beta', 'Run performance checks', 'dist/commands/perf.js', directCommand(() => import('../commands/perf.js'), 'dist/commands/perf.js')),
     bench: entry('beta', 'Run core trust-kernel benchmark budgets', 'dist/core/commands/bench-command.js', argsCommand(() => import('../core/commands/bench-command.js'), 'benchCommand', 'dist/core/commands/bench-command.js')),
-    'mcp-server': entry('beta', 'Run a stdio MCP server exposing SKS commands as tools for any MCP-capable agent host', 'dist/core/commands/mcp-server-command.js', argsCommand(() => import('../core/commands/mcp-server-command.js'), 'mcpServerCommand', 'dist/core/commands/mcp-server-command.js'), {
+    'mcp-server': entry('beta', 'Run a stdio MCP server exposing SKS commands as tools for MCP-capable agent hosts', 'dist/core/commands/mcp-server-command.js', argsCommand(() => import('../core/commands/mcp-server-command.js'), 'mcpServerCommand', 'dist/core/commands/mcp-server-command.js'), {
         skipMigrationGate: true,
         allowedDuringActiveRoute: true,
         activeRoutePolicy: 'always'

package/dist/cli/help-fast.js

@@ -8,7 +8,7 @@ SNEAKOSCOPE CODEX v${PACKAGE_VERSION}
 
   $Plan "task"              plan only; writes .sneakoscope/plans, no code edits
   $Work                     execute the latest plan with evidence gates
-  $Swarm "task"             dynamic Naruto swarm with machine verification
+  $Swarm "task"             Naruto alias for the Codex official subagent workflow
 
 Local surfaces