atris 3.38.0 → 3.41.0

package/AGENTS.md

@@ -1,9 +1,28 @@
-# AGENTS.md — Universal Agent Instructions
+# AGENTS.md: Universal Agent Instructions
 
-> **New here?** Read [`FOR_AGENTS.md`](FOR_AGENTS.md) — open letter from an agent who tried it.
+> **New here?** Read [`FOR_AGENTS.md`](FOR_AGENTS.md): open letter from an agent who tried it.
 
 > Works with: Claude Code, Cursor, Codex, OpenClaw, Windsurf, and any AI coding agent.
 
+## The way of talking (every reply, no exceptions)
+
+Talk like a person. This governs every message you send the operator: boot, status, summaries, everything.
+
+- Plain words. Say what happened and what it means for the reader, cause and effect. Not the machinery.
+- No task codes, branch names, commit hashes, PR numbers, or system nouns (worktree, verifier, projection, tick) in the message body. If the reader needs a command, put ONE copyable line at the end.
+- No insider terms without defining them in the same breath. A sharp non-engineer must follow it on first read.
+- Lead with the outcome. Complete sentences, one or two per paragraph, blank line between. No headers, bullet stacks, or tables in chat replies; those belong in files.
+- Detail lives in files and receipts. Offer "want the detail?" instead of dumping it.
+- No em dashes anywhere; use a comma, colon, or period.
+- The test before sending: read it fried at 2am. If decoding takes work, rewrite it.
+
+copy these shapes:
+- **Status:** "The reply check is built. I am running the final checks now, so the result is not ready yet."
+- **Landing:** "Replies now get a plain-language check before they reach you. The checks passed, and the change is ready."
+- **Failure:** "The plain-language check could not run because its model was unavailable. Your reply still went through, so no work was blocked."
+
+Canonical source: `atris/atris.md` `## voice`. This block is a copy for tools that only read `AGENTS.md`; if they differ, atris.md wins.
+
 ## Quick Start
 
 ```bash
@@ -31,7 +50,7 @@ task truth, proof, review, and backend/cloud sync all flow through Atris.
 |------|---------|
 | `atris/atris.md` | Protocol/backbone for this workspace |
 | `atris/PERSONA.md` | Communication style (read first) |
-| `atris task` | Current tasks, claims, dialogue, proof — deep dive: [`tasks.md`](tasks.md) |
+| `atris task` | Current tasks, claims, dialogue, proof; deep dive: [`tasks.md`](tasks.md) |
 | `.atris/state/tasks.projection.json` | Readable task projection for UIs/agents |
 | `atris/TODO.md` | Rendered/legacy task view only |
 | `atris/MAP.md` | Navigation (where is X?) |
@@ -125,13 +144,13 @@ member -> mission start --verify -> status --status active -> one bounded step -
 - Close: if the verifier passes, run `atris mission complete <id> --proof "<receipt_path>"`; if current-agent work should keep going, repeat status -> step -> tick.
 - Rollout: `atrisos-backend` and `atrisos-web` agents must check active missions before picking work; if no active mission exists and autonomy was requested, create one with owner, verifier, lane, and stop condition first.
 
-## Build Craft — what decides acceptance
+## Build Craft: what decides acceptance
 
 Mined from this repo's receipts (803 receipts, 1711 episodes): proofs naming a
 runnable verify command were accepted 670/674 at the gate; 24 of 28 bounces
 named none. These rules are the difference, in priority order:
 
-1. **Name a runnable verify command in every proof**, and run it bare — never
+1. **Name a runnable verify command in every proof**, and run it bare; never
    `test | tail` or `| grep`; a pipe replaces your exit code with the filter's.
 2. **A task naming a spec file (`atris/features/<name>/idea.md`) is a contract.**
    Read it before writing code; build the named slice only; use its verify
@@ -157,7 +176,7 @@ named none. These rules are the difference, in priority order:
    reproduces it against the real runtime, not a mock that can stay green
    through breakage.
 10. **When your engine dies mid-build (credits, limits), that is a staffing
-    event, not a failure** — leave the worktree intact with a note; the
+    event, not a failure**; leave the worktree intact with a note; the
     conductor restaffs it.
 
 ## Rules

package/atris/PERSONA.md

@@ -1,4 +1,4 @@
-# PERSONA.md — Atris Agent Personality
+# PERSONA.md: Atris Agent Personality
 
 This defines how Atris agents communicate, decide, and work.
 
@@ -28,7 +28,7 @@ This defines how Atris agents communicate, decide, and work.
 
 ## Core Workflow
 
-**Read before you act.** Before planning or building, read the relevant files. Understand the current state. Your first action in any new area is always reconnaissance — not execution.
+**Read before you act.** Before planning or building, read the relevant files. Understand the current state. Your first action in any new area is always reconnaissance, not execution.
 
 **Ask for intent when it's fuzzy or the blast radius is real.** A vague brief gets 2-3 named interpretations with a recommendation, not open questions. Small, reversible, clearly-scoped work: act, then report. Standing autonomy (autoland, push-asap) means asking permission for routine work is a failure smell, not politeness.
 
@@ -52,9 +52,13 @@ We can always add layer by layer.
 
 ## Communication Style
 
-**3-4 sentences max.** No verbose explanations. Get to the point.
+**Talk like a person.** Every message the operator reads: plain words, what happened and what it means for them, cause and effect. Not the machinery.
 
-Direct and casual tone. No corporate speak.
+**No codes in the message body.** No task ids, branch names, commit hashes, PR numbers, or system nouns (worktree, verifier, projection, tick). If the reader needs a command, ONE copyable line at the end. Insider terms get defined in the same breath or cut.
+
+**Lead with the outcome.** Complete sentences, one or two per paragraph, blank line between. No headers, bullet stacks, or tables in chat replies. Detail lives in files; offer "want the detail?" instead of dumping it.
+
+**The test before sending:** read it fried at 2am. If decoding takes work, rewrite it.
 
 If something is slop, call it out. Optimize ruthlessly.
 

package/atris.md

@@ -83,6 +83,13 @@ What you ship should not read as generated. The test: if someone said "an AI mad
 
 The same discipline for words. Output stays sharp no matter how bloated the context. A full context is not license to ramble.
 
+- **The way of talking.** Chat replies to the operator carry no task codes, branch names, commit hashes, PR numbers, or system nouns (worktree, verifier, projection, tick); if the reader needs a command, it is ONE copyable line at the end. Insider terms get defined in the same breath or cut. The test before sending: read it fried at 2am; if decoding takes work, rewrite. This is the canonical statement; `AGENTS.md` and other adapters carry a copy of it, never a variant.
+
+copy these shapes:
+- **Status:** "The reply check is built. I am running the final checks now, so the result is not ready yet."
+- **Landing:** "Replies now get a plain-language check before they reach you. The checks passed, and the change is ready."
+- **Failure:** "The plain-language check could not run because its model was unavailable. Your reply still went through, so no work was blocked."
+
 - **Lead with the move.** Answer first, support after. No preamble, no agreement reflex ("great question", "you're absolutely right").
 - **Specific over buzzy.** Name the exact thing. If you can't, you don't understand it yet; go look, don't hedge.
 - **Cut filler.** Drop "it's worth noting", "in order to", "leverage", "seamless", "robust", "delve", stacked hedges, and em dashes. `atris slop` flags the prose tells (em-dash, hype-copy) too.

package/ax

@@ -13,6 +13,7 @@ const { loadCredentials } = require('./utils/auth');
 const { apiRequestJson, getApiBaseUrl } = require('./utils/api');
 const missionRuntime = require('./lib/mission-runtime-loop');
 const { loadOverrides, pickLane } = require('./lib/ax-auto-lane');
+const { buildToolResultBody } = require('./lib/tool-result-encode');
 
 const EXIT_WORDS = new Set(['exit', 'quit', ':q']);
 const BACKEND = {
@@ -2871,7 +2872,7 @@ async function postTurn(message, options = {}) {
         if (options.business) {
           return options.business.postToolResult(callId, result, backendBaseUrl({ route: 'cloud' }));
         }
-        return postJson(BACKEND_API_TOOL_RESULT_PATH, { call_id: callId, result }, {
+        return postJson(BACKEND_API_TOOL_RESULT_PATH, buildToolResultBody(callId, result), {
           token,
           route: endpointRoute,
           timeoutMs: 30000,

package/bin/atris.js

@@ -504,7 +504,7 @@ function showHelp() {
   console.log('  activate   - Load Atris context');
   console.log('  radar      - Show live agents joined with tasks, missions, and worktrees');
   console.log('  stream     - Watch the whole team work live in one terminal');
-  console.log('  team       - Show who is awake, what they are doing, and running loops');
+  console.log('  team       - One team view: members, roles, engine assignments (presence for live)');
   console.log('  watch      - Turn one sentence into an always-on background watcher');
   console.log('  ctop       - Show a process-first live agent CPU/memory view');
   console.log('  launchpad  - Show the next action from local brain, task, mission, and proof state');
@@ -520,15 +520,17 @@ function showHelp() {
   console.log('  harvest    - Find bugs and next actions from receipts, run logs, and thinking');
   console.log('  verify     - Validate work is done (tests, MAP.md, changes)');
   console.log('  slop       - Deterministic slop detector: frontend/prose tells, dead code (slop dead --exports)');
+  console.log('  voice      - Scan binary voice tells or judge reply shape and plainness');
   console.log('  recover    - Explain stopped missions; --apply performs safe recovery only');
   console.log('  task       - Local agent task plane (atomic claims, TODO import)');
-  console.log('  golden path (zero human turns):');
+  console.log('  golden path (one tick, by cron or by hand):');
   console.log('    atris task delegate "fix the login bug" --to <member>');
   console.log('    atris task claim <id> --as <member>');
   console.log('    ... build ...');
   console.log('    atris task ready <id> --verify');
   console.log('    atris autoland tick   # second check runs, task lands');
   console.log('  mission    - Goal + loop + member owner + verifier + receipt; --budget quick|long|deep sets bounded tiers');
+  console.log('  decide     - Answer open mission asks and send the decision to the next tick');
   console.log('  release    - Tag release, bump version, create GitHub release, draft /launch');
   console.log('  learn      - Project learnings (patterns, pitfalls, preferences)');
   console.log('  study      - On-demand learning feed: ingest topic, start server, open browser');
@@ -537,6 +539,7 @@ function showHelp() {
   console.log('  avail      - Booking availability (/book/{username} weekly windows)');
   console.log('  brain      - Compile MAP/TODO/wiki/state into a loadable agent brain');
   console.log('  lesson     - Append a one-line lesson to atris/lessons.md (mine: distill receipts/episodes/scorecards into policy lessons)');
+  console.log('  taste      - Record the operator\'s keep, kill, and more creative verdicts');
   console.log('  teach      - Turn a bad turn into a failing benchmark, then a permanent guard (red gate: no promotion without a proven failure)');
   console.log('  ingest     - Local-first wiki ingest into atris/wiki/');
   console.log('  query      - Local-first wiki query against atris/wiki/');
@@ -551,7 +554,7 @@ function showHelp() {
   console.log('  worktree   - Isolated Git worktrees plus guarded ship/merge for parallel agents');
   console.log('  land       - The landing: what is actually done vs still in the air; --reap backs up + clears overdue');
   console.log('  drive      - One self-driving tick: mission doctor -> auto-fix -> count disengagements');
-  console.log('  autoland   - Approve the policy once; certified work lands itself, you keep irreversible calls');
+  console.log(`  autoland   - Approve the policy once; ${require('../lib/autoland').certifiedWorkLandsPhrase(process.cwd())}, you keep irreversible calls`);
   console.log('  engine     - Engine registry: list/resolve roles, health flips, default engine, `engine test`, and dispatch flights');
   console.log('  router     - inspect ax lane outcomes and promote gated reflex overrides');
   console.log('  sign       - Co-author trailer on every commit in an atris workspace (on/off/status)');
@@ -588,7 +591,7 @@ function showHelp() {
   console.log('  browse [query]     - Discover workspace templates');
   console.log('  fork <template>    - Clone a template into a new workspace');
   console.log('  publish            - Share your workspace as a template');
-  console.log('  pack               - Publish or install an Atris brain zip');
+  console.log('  pack               - Craft, publish, share, browse, or install an Atris brain zip');
   console.log('  sleep [business|member] [--loop id] - Pause compute or flip a member switch');
   console.log('  wake [business|member] [--loop id]  - Resume compute or flip a member switch');
   console.log('');
@@ -597,6 +600,8 @@ function showHelp() {
   console.log('  business onboard       - Onboard from sparse input (--name, --website, --contact)');
   console.log('  business add <slug>    - Connect a business');
   console.log('  business list          - Show connected businesses');
+  console.log('  business room [slug]   - Show the live room from the production API');
+  console.log('  business store on      - Enable the room store with its current catalog');
   console.log('  business remove <slug> - Disconnect a business');
   console.log('  business team [slug]   - Show members, roles, and admin access');
   console.log('  business health <slug> - Health report (members, workspace, issues)');
@@ -1815,6 +1820,18 @@ if (command === 'init') {
 } else if (command === 'router') {
   const code = require('../commands/router').routerCommand(process.argv.slice(3));
   if (Number.isInteger(code) && code !== 0) process.exit(code);
+} else if (command === 'decide') {
+  // process.exit() can outrun a piped stdout: writes beyond the 64KB pipe
+  // buffer are async, so large --json payloads truncate at 64KB multiples.
+  // Queue an empty write and exit from its callback — it fires only after
+  // every earlier buffered write has drained (BCK-1306).
+  const exitAfterStdoutDrain = (code) => {
+    if (process.stdout.writableLength === 0) process.exit(code);
+    else process.stdout.write('', () => process.exit(code));
+  };
+  Promise.resolve(require('../commands/decide').decideCommand(process.argv.slice(3)))
+    .then(() => exitAfterStdoutDrain(process.exitCode || 0))
+    .catch((err) => { console.error(`\n✗ Error: ${err.message || err}`); exitAfterStdoutDrain(1); });
 } else if (command === 'mission') {
   // process.exit() can outrun a piped stdout: writes beyond the 64KB pipe
   // buffer are async, so large --json payloads truncate at 64KB multiples.
@@ -1962,7 +1979,7 @@ if (command === 'init') {
   if (logsArgs.includes('--help') || logsArgs.includes('-h') || logsArgs[0] === 'help') {
     console.log('Usage: atris logs [--date YYYY-MM-DD] [--json]');
     console.log('');
-    console.log('  atris logs               Show today\'s workspace + team journal digest');
+    console.log('  atris logs               Show today\'s journal entries, completed receipts, and inbox in one view');
     console.log('  atris logs --date <day>  Show the digest for a specific day');
     console.log('  atris logs --json        Emit the digest as JSON');
     process.exit(0);
@@ -2633,6 +2650,10 @@ if (command === 'init') {
   const subcommand = process.argv[3];
   const args = process.argv.slice(4);
   require('../commands/lesson')(subcommand, ...args);
+} else if (command === 'taste') {
+  Promise.resolve(require('../commands/taste').tasteCommand(process.argv.slice(3)))
+    .then((code) => process.exit(typeof code === 'number' ? code : 0))
+    .catch((err) => { console.error(`taste could not continue: ${err.message || err}.`); process.exit(1); });
 } else if (command === 'teach') {
   const subcommand = process.argv[3];
   const args = process.argv.slice(4);
@@ -2711,7 +2732,7 @@ if (command === 'init') {
   const subcommand = process.argv[3];
   const args = process.argv.slice(4);
   require('../commands/business').businessCommand(subcommand, ...args)
-    .then(() => process.exit(0))
+    .then(() => process.exit(process.exitCode || 0))
     .catch((err) => { console.error(`\n✗ Error: ${err.message || err}`); process.exit(1); });
 } else if (command === 'soul') {
   const args = process.argv.slice(3);
@@ -2806,6 +2827,10 @@ if (command === 'init') {
   Promise.resolve(require('../commands/slop').slopCommand(process.argv.slice(3)))
     .then((code) => process.exit(typeof code === 'number' ? code : 0))
     .catch((err) => { console.error(`\n✗ Error: ${err.message || err}`); process.exit(1); });
+} else if (command === 'voice') {
+  Promise.resolve(require('../commands/voice').voiceCommand(process.argv.slice(3)))
+    .then((code) => process.exit(typeof code === 'number' ? code : 0))
+    .catch((err) => { console.error(`\n✗ Error: ${err.message || err}`); process.exit(1); });
 } else if (command === 'write') {
   // Write: guided writing sessions — human types every word, atris structures + reviews (plan-do-review for prose).
   Promise.resolve(require('../commands/write').writeCommand(process.argv.slice(3)))

package/commands/agent-spawn.js

@@ -2,6 +2,7 @@ const fs = require('fs');
 const path = require('path');
 const crypto = require('crypto');
 const { spawnSync } = require('child_process');
+const { hasFlag: hasSingleFlag } = require('../lib/arg-parser');
 
 const SPAWN_SCHEMA = 'atris.agent_spawn.v1';
 const DOGFOOD_SCHEMA = 'atris.agent_cli_dogfood.v1';
@@ -63,8 +64,9 @@ function flagValue(args, names) {
   return null;
 }
 
-function hasFlag(args, names) {
-  return names.some(name => args.includes(name));
+// This command accepts alias lists, while the shared helper checks one name.
+function hasAnyFlag(args, names) {
+  return names.some(name => hasSingleFlag(args, name));
 }
 
 function positionalArgs(args) {
@@ -148,11 +150,11 @@ function commandForEngine(request) {
 }
 
 function parseSpawnArgs(args = []) {
-  const help = args.length === 0 || hasFlag(args, ['--help', '-h']) || args[0] === 'help';
+  const help = args.length === 0 || hasAnyFlag(args, ['--help', '-h']) || args[0] === 'help';
   if (help) return { help: true };
 
-  const json = hasFlag(args, ['--json']);
-  const dryRun = hasFlag(args, ['--dry-run']);
+  const json = hasAnyFlag(args, ['--json']);
+  const dryRun = hasAnyFlag(args, ['--dry-run']);
   const roleFlag = flagValue(args, ['--role']);
   const engine = String(flagValue(args, ['--engine']) || 'manual').toLowerCase();
   if (!ALLOWED_ENGINES.has(engine)) throw new Error(`Unknown engine: ${engine}`);
@@ -209,10 +211,10 @@ function showSpawnHelp(output = console.log, commandName = 'atris agent spawn')
 }
 
 function parseDogfoodArgs(args = []) {
-  const help = hasFlag(args, ['--help', '-h']) || args[0] === 'help';
-  const json = hasFlag(args, ['--json']);
-  const live = hasFlag(args, ['--live']);
-  const noWrite = hasFlag(args, ['--no-write']);
+  const help = hasAnyFlag(args, ['--help', '-h']) || args[0] === 'help';
+  const json = hasAnyFlag(args, ['--json']);
+  const live = hasAnyFlag(args, ['--live']);
+  const noWrite = hasAnyFlag(args, ['--no-write']);
   const model = flagValue(args, ['--model']) || 'glm-5.2';
   const timeoutRaw = Number(flagValue(args, ['--timeout']) || 45);
   const timeoutMs = Math.max(5, Math.min(300, timeoutRaw)) * 1000;
@@ -431,7 +433,7 @@ function agentSpawnCommand(args = [], deps = {}) {
 function agentSpawnListCommand(args = [], deps = {}) {
   const root = deps.root || process.cwd();
   const output = deps.output || ((line = '') => console.log(line));
-  const json = hasFlag(args, ['--json']);
+  const json = hasAnyFlag(args, ['--json']);
   const requests = loadSpawnRequests(root).reverse();
   const payload = { ok: true, action: 'spawn_list', requests, state_path: spawnStatePath(root) };
   if (json) {
@@ -451,7 +453,7 @@ function agentSpawnListCommand(args = [], deps = {}) {
 function agentSpawnStatusCommand(args = [], deps = {}) {
   const root = deps.root || process.cwd();
   const output = deps.output || ((line = '') => console.log(line));
-  const json = hasFlag(args, ['--json']);
+  const json = hasAnyFlag(args, ['--json']);
   const id = positionalArgs(args)[0];
   if (!id) throw new Error('Missing spawn id. Usage: atris agent spawn-status <id>');
   const request = loadSpawnRequests(root).find(req => req.id === id);

package/commands/autoland.js

@@ -6,7 +6,7 @@ const os = require('os');
 const { spawnSync } = require('child_process');
 
 const autoland = require('../lib/autoland');
-const { gateForHuman } = require('../lib/voice-gate');
+const { gateForHuman, plainLandingReason } = require('../lib/voice-gate');
 const { evaluateAutoAccept } = require('../lib/auto-accept-certified');
 const { operatorReady, hasAgentJargon, explainResult } = autoland;
 const MISSION_AUTO_VERIFY_STATUSES = new Set(['planning', 'paused', 'ready']);
@@ -436,26 +436,9 @@ function protectedReviewWaiting(root, policy = {}, waiting = []) {
     .filter((row) => !alreadyWaiting.has(row.ref));
 }
 
-function plainReason(reason) {
-  const map = {
-    denied_tag_billing: 'money: human decision required',
-    denied_tag_deploy: 'a deploy: human decision required',
-    denied_tag_security: 'security: human decision required',
-    denied_tag_customer: 'customer-facing: human decision required',
-    denied_tag_external: 'outward-facing: human decision required',
-    denied_tag_feedback: 'customer feedback: human decision required',
-    needs_second_reviewer_or_third_pass: 'needs one more independent check first',
-    needs_independent_reviewer: 'built and judged by the same actor, needs an independent check',
-    verifier_is_builder: 'the re-check actor built this row, another actor must re-check',
-    judge_equals_worker: 'built and judged by the same actor, hand the review to someone else',
-    not_agent_certified: 'not certified yet',
-    insufficient_review_passes: 'not enough review passes yet',
-    strict_verify_missing: 'no recorded check command to re-run',
-    verify_failed: 'its check command failed on re-run',
-    proof_unmerged_or_draft_pr_boundary: 'its proof points at an unmerged draft',
-  };
-  return map[reason] || reason.replace(/_/g, ' ');
-}
+// One reason table for every human-facing landing refusal lives in
+// lib/voice-gate.js (plainLandingReason); this alias keeps call sites short.
+const plainReason = plainLandingReason;
 
 function showStatus(root, args) {
   const json = args.includes('--json');
@@ -475,7 +458,8 @@ function showStatus(root, args) {
   // running hourly for weeks, and would keep reporting "running" for one
   // someone deleted. A loop that cannot confirm its own liveness from
   // evidence cannot be trusted to say it is healthy.
-  const heartbeatInstalled = heartbeatLiveness(root, policy);
+  const heartbeatInstalled = autoland.heartbeatLiveness(root, policy);
+  const heartbeatLive = autoland.heartbeatIsLive(root);
 
   if (json) {
     console.log(JSON.stringify({
@@ -491,7 +475,7 @@ function showStatus(root, args) {
   }
 
   console.log('');
-  console.log('autoland: certified work lands itself; you keep the irreversible calls');
+  console.log(`autoland: ${autoland.certifiedWorkLandsPhrase(root)}; you keep the irreversible calls`);
   console.log('');
   const policyOwner = String(policy?.enabled_by || 'unknown').trim() || 'unknown';
   const policyText = enabled
@@ -501,26 +485,33 @@ function showStatus(root, args) {
     : 'off - everything waits for you';
   console.log(`  policy: ${policyText}`);
   if (enabled && acceptAll) console.log('  bar: everything lands except the protected lanes (money, deploys, security, customer, outward)');
-  console.log(`  heartbeat: ${heartbeatStatusText(root, policy)}`);
+  console.log(`  heartbeat: ${autoland.heartbeatStatusText(root, policy)}`);
   if (policy && policy.imessage_to) console.log(`  daily message: ${policy.imessage_to} at ${policy.digest_hour ?? autoland.DEFAULT_DIGEST_HOUR}:00`);
   const reapTrouble = autoland.readState(root).last_reap_error;
   if (reapTrouble) console.log(`  cleanup trouble: landing sweep failed on ${reapTrouble.date} (${reapTrouble.error}) - run: atris land --reap`);
   console.log('');
-  if (readyForRecheck.length > 0) {
-    console.log(`  ready for heartbeat recheck: ${readyForRecheck.length}`);
-    for (const r of readyForRecheck.slice(0, 10)) console.log(`    rechecks then lands  ${r.ref}`);
-  } else {
-    console.log('  nothing is ready to land on its own right now.');
-  }
   const humanOnly = blocked.filter((r) => String(r.reason || '').startsWith('denied_tag_'));
   const needsWork = blocked.filter((r) => !String(r.reason || '').startsWith('denied_tag_'));
   if (humanOnly.length > 0) {
-    console.log(`  protected reviews waiting on you: ${humanOnly.length}`);
-    for (const r of humanOnly.slice(0, 10)) console.log(`    waits for you  ${r.ref} - ${plainReason(r.reason)}`);
+    console.log(`  yours to decide first: ${humanOnly.length} protected ${humanOnly.length === 1 ? 'review' : 'reviews'}.`);
+    for (const r of humanOnly.slice(0, 10)) console.log(`    ${r.ref} waits for you; ${plainReason(r.reason)}.`);
+  }
+  if (readyForRecheck.length > 0) {
+    if (heartbeatLive) {
+      console.log(`  landing on their own after the hourly recheck: ${readyForRecheck.length}.`);
+      for (const r of readyForRecheck.slice(0, 10)) console.log(`    ${r.ref} passes the recheck, then lands itself.`);
+    } else {
+      console.log(`  ready to land ${autoland.STALE_HEARTBEAT_LANDING}: ${readyForRecheck.length}.`);
+      for (const r of readyForRecheck.slice(0, 10)) {
+        console.log(`    ${r.ref} passes the recheck, then lands ${autoland.STALE_HEARTBEAT_LANDING}.`);
+      }
+    }
+  } else {
+    console.log('  nothing is ready to land on its own right now.');
   }
   if (needsWork.length > 0) {
-    console.log(`  not ready yet: ${needsWork.length}`);
-    for (const r of needsWork.slice(0, 10)) console.log(`    held back      ${r.ref} - ${plainReason(String(r.reason || ''))}`);
+    console.log(`  still needs work before it can land: ${needsWork.length}.`);
+    for (const r of needsWork.slice(0, 10)) console.log(`    ${r.ref} stays put; ${plainReason(String(r.reason || ''))}.`);
   }
   if (waiting.length > 0) {
     console.log('');
@@ -699,48 +690,8 @@ function persistTickReceipt(root, receipt) {
 // loud, and call out the two states that mean the loop is wedged rather than
 // idle: work blocked behind a check the harness could not execute, and work
 // that passed every gate and still did not land.
-// Installed and alive are different claims, and only one of them is worth
-// printing. Deliberately does NOT shell out to crontab: `autoland status` is
-// read-only and must stay fast (a stalled crontab would block it for the full
-// 10s read timeout — see 'autoland status never probes a stalled crontab').
-// Tick receipts are local file reads and are stronger evidence anyway: a
-// crontab line proves configuration, a recent receipt proves the loop ran.
-function heartbeatLiveness(root, policy) {
-  return typeof policy?.heartbeat_installed === 'boolean' ? policy.heartbeat_installed : null;
-}
-
-function lastTickAgeHours(root) {
-  try {
-    const runsDir = path.join(root, 'atris', 'runs');
-    const newest = fs.readdirSync(runsDir)
-      .filter((f) => f.startsWith('autoland-tick-') && f.endsWith('.json'))
-      .sort()
-      .pop();
-    if (!newest) return null;
-    const stamp = fs.statSync(path.join(runsDir, newest)).mtimeMs;
-    return (Date.now() - stamp) / 3_600_000;
-  } catch {
-    return null;
-  }
-}
-
-function heartbeatStatusText(root, policy) {
-  const ageHours = lastTickAgeHours(root);
-  const ageText = ageHours === null
-    ? null
-    : ageHours < 1 ? 'under an hour' : `${Math.floor(ageHours)}h`;
-  // Evidence first: a receipt from the last couple of hours proves the loop is
-  // alive no matter what the policy file remembers. The hourly cron makes two
-  // missed hours an outage rather than jitter — and a heartbeat that has gone
-  // quiet is the single most useful thing this line can say, because that is
-  // the state nobody notices.
-  if (ageHours !== null && ageHours <= 2) return `running hourly (last tick ${ageText} ago)`;
-  if (ageHours !== null) return `SILENT - last tick ${ageText} ago; run atris autoland tick`;
-  const installed = heartbeatLiveness(root, policy);
-  if (installed === false) return 'not installed - run atris autoland on';
-  if (installed === true) return 'installed, but no tick has ever run - run atris autoland tick';
-  return 'unknown - run atris autoland on to check and repair';
-}
+// Heartbeat liveness helpers live in lib/autoland.js so task ready copy and
+// status share one gate (recent tick receipt within two hours).
 
 function stuckBacklogNote(receipt) {
   const parts = [];
@@ -1049,7 +1000,7 @@ function runTickBody(root, { json, policy, receipt }) {
 
 function showHelp() {
   console.log('');
-  console.log('atris autoland: you approve the policy once; certified work lands itself');
+  console.log(`atris autoland: you approve the policy once; ${autoland.certifiedWorkLandsPhrase(process.cwd())}`);
   console.log('');
   console.log('finished work that passed its checks and two independent reviews lands');
   console.log('automatically with a receipt. money, deploys, security, customer, and');

package/commands/bench.js

@@ -8,12 +8,10 @@ const {
   taskMetadata,
 } = require('../lib/bench/runner');
 const { buildBenchReport, renderBenchReportText } = require('../lib/bench/report');
+const { hasFlag } = require('../lib/arg-parser');
 
-function hasFlag(args, name) {
-  return args.includes(name);
-}
-
-function readFlag(args, name) {
+// Preserve the existing rule that any inline value wins over a split value.
+function readInlineFirstFlag(args, name) {
   const prefix = `${name}=`;
   const inline = args.find((arg) => String(arg).startsWith(prefix));
   if (inline) return String(inline).slice(prefix.length);
@@ -58,11 +56,11 @@ async function runCommand(args) {
   const asJson = hasFlag(args, '--json');
   try {
     const result = await runBench({
-      pack: readFlag(args, '--pack') || undefined,
-      engine: readFlag(args, '--engine') || undefined,
+      pack: readInlineFirstFlag(args, '--pack') || undefined,
+      engine: readInlineFirstFlag(args, '--engine') || undefined,
       taskIds: readRepeatedFlag(args, '--task'),
-      label: readFlag(args, '--label'),
-      experiment: readFlag(args, '--experiment'),
+      label: readInlineFirstFlag(args, '--label'),
+      experiment: readInlineFirstFlag(args, '--experiment'),
       updateBaseline: hasFlag(args, '--update-baseline'),
       stateRoot: process.cwd(),
     });
@@ -87,7 +85,7 @@ async function runCommand(args) {
 }
 
 function tasksCommand(args) {
-  const pack = readFlag(args, '--pack') || undefined;
+  const pack = readInlineFirstFlag(args, '--pack') || undefined;
   const tasks = taskMetadata({ pack });
   if (hasFlag(args, '--json')) {
     console.log(JSON.stringify({
@@ -121,7 +119,7 @@ function packsCommand(args) {
 }
 
 function resultsCommand(args) {
-  const last = Number(readFlag(args, '--last') || 0);
+  const last = Number(readInlineFirstFlag(args, '--last') || 0);
   const results = readResultRecords({ stateRoot: process.cwd(), last });
   if (hasFlag(args, '--json')) {
     console.log(JSON.stringify({
@@ -143,7 +141,7 @@ function resultsCommand(args) {
 }
 
 function reportCommand(args) {
-  const pack = readFlag(args, '--pack') || undefined;
+  const pack = readInlineFirstFlag(args, '--pack') || undefined;
   const asJson = hasFlag(args, '--json');
   try {
     const report = buildBenchReport({ pack, stateRoot: process.cwd() });