Grow implementation kit: 22 patterns, 22 contracts, 671 resources

README.de.md

@@ -28,7 +28,7 @@ Ein Loop entdeckt Arbeit, übergibt sie an einen oder mehrere Agenten, prüft da
 
 Loop Engineering umfasst wiederkehrende KI-Agenten- und Coding-Agent-Systeme mit expliziten Auslösern, Verifikation und dauerhaftem Zustand. Software-Event-Loops, Regelungstechnik, Growth Loops, generische Automatisierung und nicht-KI-bezogene Feedback-Loops gehören nicht dazu.
 
-Starte mit 669 Ressourcen, 20 operativen Loop-Patterns, 20 schemavalidierten Loop-Contracts und 8 Runtime-Startern (3 ausführbare Programme und 5 anpassbare Vorlagen), ergänzt durch eine Community-Galerie und 8 Sprachen.
+Starte mit 671 Ressourcen, 20 operativen Loop-Patterns, 20 schemavalidierten Loop-Contracts und 8 Runtime-Startern (3 ausführbare Programme und 5 anpassbare Vorlagen), ergänzt durch eine Community-Galerie und 8 Sprachen.
 
 ## Mentales Modell
 

README.es.md

@@ -28,7 +28,7 @@ Un loop descubre trabajo, lo delega a uno o más agentes, verifica el resultado,
 
 Loop Engineering abarca sistemas recurrentes de agentes de IA y coding agents con disparadores explícitos, verificación y estado duradero. No incluye event loops de software, teoría de control, growth loops, automatización genérica ni feedback loops no relacionados con IA.
 
-Empieza con 669 recursos, 20 patrones operativos, 20 loop contracts validados por esquema y 8 puntos de partida de runtime (3 ejecutables y 5 plantillas listas para adaptar), además de una galería comunitaria y 8 idiomas.
+Empieza con 671 recursos, 20 patrones operativos, 22 loop contracts validados por esquema y 8 puntos de partida de runtime (3 ejecutables y 5 plantillas listas para adaptar), además de una galería comunitaria y 8 idiomas.
 
 ## Modelo Mental
 

README.fr.md

@@ -28,7 +28,7 @@ Un loop découvre du travail, le délègue à un ou plusieurs agents, vérifie l
 
 Loop Engineering couvre les systèmes récurrents d'agents IA et de coding agents avec des déclencheurs explicites, une vérification et un état durable. Il n'inclut ni les event loops logiciels, ni la théorie du contrôle, ni les growth loops, ni l'automatisation générique, ni les feedback loops sans rapport avec l'IA.
 
-Commencez avec 669 ressources, 20 patterns opérationnels, 20 loop contracts validés par schéma et 8 points de départ d'exécution (3 exécutables et 5 modèles prêts à adapter), ainsi qu'une galerie communautaire et 8 langues.
+Commencez avec 671 ressources, 22 patterns opérationnels, 22 loop contracts validés par schéma et 8 points de départ d'exécution (3 exécutables et 5 modèles prêts à adapter), ainsi qu'une galerie communautaire et 8 langues.
 
 ## Modèle Mental
 

README.ja.md

@@ -28,7 +28,7 @@ Loop は作業を発見し、1 つ以上の agents に委任し、結果を確
 
 Loop Engineering は、明示的なトリガー、検証、永続状態を備えた反復型 AI agent / coding agent システムを対象とします。software event loop、制御理論、growth loop、一般的な workflow automation、非 AI の feedback loop は含みません。
 
-669 件の監査済みリソース、20 個の運用パターン、20 個の schema 検証済み loop contract、8 個の runtime starter（3 個の実行ファイルと 5 個の適応可能なテンプレート）から始められます。コミュニティ gallery と 8 言語の導入も利用できます。
+671 件の監査済みリソース、20 個の運用パターン、20 個の schema 検証済み loop contract、8 個の runtime starter（3 個の実行ファイルと 5 個の適応可能なテンプレート）から始められます。コミュニティ gallery と 8 言語の導入も利用できます。
 
 ## メンタルモデル
 

README.ko.md

@@ -28,7 +28,7 @@ Loop 는 작업을 발견하고, 하나 이상의 agents 에게 위임하고,
 
 Loop Engineering 은 명시적 트리거, 검증, 지속 상태를 갖춘 반복 AI agent / coding agent 시스템을 다룹니다. software event loop, control theory, growth loop, 일반적인 workflow automation, 비 AI feedback loop 는 포함하지 않습니다.
 
-669개의 감사된 리소스, 20개의 운영 loop 패턴, 20개의 스키마 검증 loop contract, 8개의 runtime starter(실행 파일 3개와 적용 가능한 템플릿 5개)로 바로 시작할 수 있습니다. 커뮤니티 gallery 와 8개 언어 안내도 제공합니다.
+671개의 감사된 리소스, 20개의 운영 loop 패턴, 20개의 스키마 검증 loop contract, 8개의 runtime starter(실행 파일 3개와 적용 가능한 템플릿 5개)로 바로 시작할 수 있습니다. 커뮤니티 gallery 와 8개 언어 안내도 제공합니다.
 
 ## 멘탈 모델
 

README.pt-BR.md

@@ -28,7 +28,7 @@ Um loop descobre trabalho, delega para um ou mais agentes, verifica o resultado,
 
 Loop Engineering abrange sistemas recorrentes de AI agents e coding agents com gatilhos explícitos, verificação e estado durável. Não inclui event loops de software, teoria de controle, growth loops, automação genérica nem feedback loops sem relação com IA.
 
-Comece com 669 recursos, 20 patterns operacionais, 20 loop contracts validados por schema e 8 starters de runtime (3 executáveis e 5 templates prontos para adaptação), além de uma galeria comunitária e 8 idiomas.
+Comece com 671 recursos, 22 patterns operacionais, 22 loop contracts validados por schema e 8 starters de runtime (3 executáveis e 5 templates prontos para adaptação), além de uma galeria comunitária e 8 idiomas.
 
 ## Modelo Mental
 

README.zh-CN.md

@@ -32,7 +32,7 @@ Prompt engineering 改进你对模型说什么。Context engineering 改进模
 
 Loop Engineering 专指具备明确触发、外部验证和持久状态的可重复 AI-agent 与 coding-agent 系统，不包括软件事件循环、控制论、增长循环、通用 workflow automation 或非 AI feedback loop。
 
-可直接从 669 篇论文、官方文档、工具、benchmark 与实践指南，20 个操作模式、20 个经 schema 校验的 loop contracts 和 8 个 runtime starters（3 个可直接执行，5 个可复制改造）开始，并通过社区 gallery 和 8 种语言入口继续探索。
+可直接从 671 篇论文、官方文档、工具、benchmark 与实践指南，20 个操作模式、20 个经 schema 校验的 loop contracts 和 8 个 runtime starters（3 个可直接执行，5 个可复制改造）开始，并通过社区 gallery 和 8 种语言入口继续探索。
 
 ## 一句话定位
 

data/README.md

@@ -1,6 +1,6 @@
 # Dataset Exports
 
-Download all 669 resources as deterministic tabular exports.
+Download all 671 resources as deterministic tabular exports.
 
 - `resources.csv` - Tabular export for spreadsheets and ad hoc analysis.
 - `resources.jsonl` - JSON Lines export and source for the Hugging Face Parquet build.

data/first_seen.json

@@ -181,5 +181,7 @@
   "https://www.starfleetmath.com/": "2026-07-22",
   "https://www.ayush.digital/blog/the-memory-heist": "2026-07-22",
   "https://mindgard.ai/blog/cursor-0day-when-full-disclosure-becomes-the-only-protection-left": "2026-07-22",
-  "https://lucumr.pocoo.org/2026/7/13/the-tower-keeps-rising/": "2026-07-22"
+  "https://lucumr.pocoo.org/2026/7/13/the-tower-keeps-rising/": "2026-07-22",
+  "patterns/agent-memory-lifecycle-loop.md": "2026-07-22",
+  "patterns/fleet-coordination-loop.md": "2026-07-22"
 }

design-qa.md

@@ -41,7 +41,7 @@
 8. Compared the published static hero and the Three.js-enhanced hero at the same 1280 x 720 viewport. The enhancement keeps the existing identity, copy, actions, spacing, and section transition while adding a restrained evidence orbit, control-plane stack, verification gate, and six lifecycle nodes behind the content.
 9. Moved the control-plane stack and verification gate outward after the first 3D pass so they frame the copy rather than crossing its reading path. Rechecked desktop and mobile rendering after the adjustment.
 10. Verified progressive enhancement behavior: the semantic hero remains complete without WebGL, forced-colors and print hide the scene, reduced-motion renders a static frame, and the animation stops when the hero leaves the viewport.
-11. Expanded the implementation kit to 20 patterns, 20 schema-checked contracts, and eight runtime starters. The runtime section now labels all five adaptable templates and all three executables, with use-case copy and an even desktop grid.
+11. Expanded the implementation kit to 22 patterns, 22 schema-checked contracts, and eight runtime starters. The runtime section now labels all five adaptable templates and all three executables, with use-case copy and an even desktop grid.
 12. Rechecked the expanded pattern and runtime sections at 1440 x 1000 and 390 x 844. Anchor offsets clear the sticky header, the page width stays equal to the viewport, and headings, icons, labels, and descriptions remain aligned.
 13. Added the role-based future-directions section and rechecked it at both target viewports. Resource Atlas expansion initially displaced direct links to later sections; post-load hash restoration now leaves `#future` 84 px below the desktop viewport top and 74 px below it on mobile.
 
@@ -76,9 +76,9 @@
 - Tablet: 768 x 900. The first pass clipped the intake and decision stations. The responsive camera now fits an 11.2-unit horizontal view between 720 px and 900 px, keeping all six stations, the retry path, and the outcome node inside the canvas.
 - Phone: 390 x 844, no horizontal overflow. The compact loop keeps all six stations, the moving work packet, legend, outcome key, and stage selector visible without crossing the hero copy.
 - Mobile navigation: opens all nine project links, reports `aria-expanded="true"`, closes on Escape, and restores focus to the menu button.
-- Pattern filter: `accessibility` returns exactly one of the 20 patterns.
-- Resource Atlas: remains idle above the fold, loads on approach or a direct resource anchor, and reports `669 of 669 resources | showing 8` on a fresh mobile load.
-- Resource interactions: the Model loop-layer filter returns 29 of 669 resources (28 papers plus Awesome Loop Models), paginates eight at a time on mobile, and renders Loopie and LoopWM with their 2026 arXiv records and adjacent-scope labels.
+- Pattern filter: `accessibility` returns exactly one of the 22 patterns.
+- Resource Atlas: remains idle above the fold, loads on approach or a direct resource anchor, and reports `669 of 671 resources | showing 8` on a fresh mobile load.
+- Resource interactions: the Model loop-layer filter returns 29 of 671 resources (28 papers plus Awesome Loop Models), paginates eight at a time on mobile, and renders Loopie and LoopWM with their 2026 arXiv records and adjacent-scope labels.
 - Resource row schema: desktop exposes Work, Year, Published at, and Evidence; mobile stacks the same labels. A `ReasoningBank` query returns one row with its ICLR venue, arXiv identifier, Tier A evidence, task fit, and source record without horizontal overflow.
 - Scene motion: the canvas frame counter advanced from 640 to 662 over 900 ms. Two canvas crops taken 1.2 seconds apart changed 17.79% of sampled RGB channels.
 - Scene pixels: the current desktop canvas has 7,062 colors with RGB standard deviations of 35.93, 29.76, and 22.29; mobile has 5,755 colors with deviations of 40.73, 30.54, and 19.70. Both renders are nonblank and well contrasted.
@@ -86,8 +86,8 @@
 - Future directions: direct-anchor restoration waits for the Resource Atlas reflow, all three audience tracks remain readable, and the agenda CTA is fully visible at desktop and mobile widths.
 - Console: no page-origin errors or warnings during mobile navigation, pattern filtering, and Resource Atlas interactions.
 - Narrow phone: 320 x 844, document width equals viewport width; the headline stays within 16 px gutters and the Three.js canvas remains fully inside the 288 px content width.
-- Source audit: 669 rows checked; 619 public sources reachable, 1 access-restricted, 49 repository-native, and 0 broken or unreachable as of 2026-07-22 UTC.
-- Data and copy: the website, README, translations, social preview source, release copy, and Hugging Face card all report 669 resources, 20 patterns, 20 contracts, 8 runtime starters, and version 0.9.0. The dataset exposes 50 fields, including `loop_layer` and `scope_fit`.
+- Source audit: 669 rows checked; 618 public sources reachable, 2 access-restricted, 51 repository-native, and 0 broken or unreachable as of 2026-07-22 UTC.
+- Data and copy: the website, README, translations, social preview source, release copy, and Hugging Face card all report 671 resources, 22 patterns, 22 contracts, 8 runtime starters, and version 0.9.0. The dataset exposes 50 fields, including `loop_layer` and `scope_fit`.
 - Semantic scene pass: the desktop hero crop paints 14.61% of pixels beyond the paper background with a 254.71 luminance range; the desktop stack paints 27.83% with a 249.98 range; the mobile hero paints 20.84% with a 229.45 range.
 - Stage interaction: selecting Verify updates `aria-pressed` across all six controls and replaces the live detail with the independent evidence-gate explanation.
 - Progressive enhancement: desktop loads the agent-loop and engineering-stack canvases; mobile loads only the agent-loop canvas and exposes the four-layer text stack. Both target viewports have zero horizontal overflow and no console messages.

docs/index.html

@@ -4,7 +4,7 @@
     <meta charset="utf-8">
     <meta name="viewport" content="width=device-width, initial-scale=1">
     <title>Awesome Loop Engineering</title>
-    <meta name="description" content="Explore 669 resources from model recurrence to governed agent operations, then build with 20 patterns, 20 contracts, and 8 runtime starters.">
+    <meta name="description" content="Explore 671 resources from model recurrence to governed agent operations, then build with 22 patterns, 22 contracts, and 8 runtime starters.">
     <meta name="robots" content="index,follow">
     <meta name="author" content="Chaoyue He">
     <link rel="canonical" href="https://chaoyue0307.github.io/awesome-loop-engineering/">
@@ -15,16 +15,16 @@
     <link href="https://fonts.googleapis.com/css2?family=IBM+Plex+Mono:wght@500;600&amp;family=Inter:wght@400;500;600;700;800&amp;display=swap" rel="stylesheet">
     <meta property="og:type" content="website">
     <meta property="og:title" content="Awesome Loop Engineering">
-    <meta property="og:description" content="Explore 669 papers, docs, tools, benchmarks, and guides, then build recurring agent systems with 20 patterns, 20 contracts, and 8 starters.">
+    <meta property="og:description" content="Explore 669 papers, docs, tools, benchmarks, and guides, then build recurring agent systems with 22 patterns, 22 contracts, and 8 starters.">
     <meta property="og:url" content="https://chaoyue0307.github.io/awesome-loop-engineering/">
-    <meta property="og:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-669-v10.png">
+    <meta property="og:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-671-v10.png">
     <meta property="og:image:width" content="1280">
     <meta property="og:image:height" content="640">
     <meta property="og:image:alt" content="Awesome Loop Engineering: design loops, not just prompts.">
     <meta name="twitter:card" content="summary_large_image">
     <meta name="twitter:title" content="Awesome Loop Engineering">
-    <meta name="twitter:description" content="669 resources from model recurrence to agent operations, plus 20 patterns, 20 contracts, and 8 starters for building reliable loops.">
-    <meta name="twitter:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-669-v10.png">
+    <meta name="twitter:description" content="671 resources from model recurrence to agent operations, plus 22 patterns, 22 contracts, and 8 starters for building reliable loops.">
+    <meta name="twitter:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-671-v10.png">
     <link rel="alternate" type="application/atom+xml" title="Awesome Loop Engineering releases" href="https://github.com/ChaoYue0307/awesome-loop-engineering/releases.atom">
     <link rel="alternate" type="application/atom+xml" title="Awesome Loop Engineering commits" href="https://github.com/ChaoYue0307/awesome-loop-engineering/commits/main.atom">
     <style>
@@ -1307,7 +1307,7 @@
           "https://huggingface.co/datasets/cy0307/awesome-loop-engineering"
         ],
         "license": "https://creativecommons.org/publicdomain/zero/1.0/",
-        "image": "https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-669-v10.png",
+        "image": "https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-671-v10.png",
         "logo": "https://chaoyue0307.github.io/awesome-loop-engineering/assets/awesome-loop-engineering-logo.png",
         "author": { "@type": "Person", "name": "Chaoyue He" }
       }
@@ -1355,7 +1355,7 @@
         "@context": "https://schema.org",
         "@type": "Dataset",
         "name": "Awesome Loop Engineering resources",
-        "description": "Explore 669 resources across model, agent, harness, workflow, operations, and evaluation layers, with publication, contribution, novelty, impact, and evidence fields.",
+        "description": "Explore 671 resources across model, agent, harness, workflow, operations, and evaluation layers, with publication, contribution, novelty, impact, and evidence fields.",
         "url": "https://chaoyue0307.github.io/awesome-loop-engineering/",
         "sameAs": [
           "https://huggingface.co/datasets/cy0307/awesome-loop-engineering",
@@ -1430,7 +1430,7 @@
             <p class="section-label hero-kicker"><i class="icon" data-lucide="repeat-2" aria-hidden="true"></i><span>Field guide for recurring agent systems</span></p>
             <h1>Awesome Loop Engineering</h1>
             <p class="tagline">Design loops, not just prompts.</p>
-            <p class="lead">Explore 669 resources from model recurrence to agent operations, then build with 20 patterns, 20 adaptable contracts, and 8 runtime starters.</p>
+            <p class="lead">Explore 671 resources from model recurrence to agent operations, then build with 22 patterns, 22 adaptable contracts, and 8 runtime starters.</p>
             <div class="actions">
               <a class="button primary" href="https://github.com/ChaoYue0307/awesome-loop-engineering#readme"><i class="icon" data-lucide="book-open" aria-hidden="true"></i>Explore resources</a>
               <a class="button" href="#choose"><i class="icon" data-lucide="compass" aria-hidden="true"></i>Choose a loop</a>
@@ -1493,7 +1493,7 @@
             </div>
           </figure>
           <div class="stats" role="list" aria-label="Field guide at a glance">
-            <div class="stat" role="listitem"><b><i class="icon" data-lucide="library" aria-hidden="true"></i>669</b><span>resources</span></div>
+            <div class="stat" role="listitem"><b><i class="icon" data-lucide="library" aria-hidden="true"></i>671</b><span>resources</span></div>
             <div class="stat" role="listitem"><b><i class="icon" data-lucide="blocks" aria-hidden="true"></i>20</b><span>operational patterns</span></div>
             <div class="stat" role="listitem"><b><i class="icon" data-lucide="file-check-2" aria-hidden="true"></i>20</b><span>loop contracts</span></div>
             <div class="stat" role="listitem"><b><i class="icon" data-lucide="play" aria-hidden="true"></i>8</b><span>runtime starters</span></div>
@@ -1793,9 +1793,9 @@
               <p class="section-label"><i class="icon" data-lucide="library" aria-hidden="true"></i><span>06 / Resource atlas</span></p>
               <h2>Find evidence, not just links.</h2>
             </div>
-            <p class="section-intro">Filter 669 resources by goal, loop layer, lifecycle, artifact type, and evidence class. Open any result to see its contribution, novelty, impact, scope, publication details, and original source.</p>
+            <p class="section-intro">Filter 671 resources by goal, loop layer, lifecycle, artifact type, and evidence class. Open any result to see its contribution, novelty, impact, scope, publication details, and original source.</p>
           </div>
-          <p class="audit-note"><strong>Last source check, 2026-07-21 UTC:</strong> 619 public links opened successfully, 1 required access, 49 point to files in this repository, and none were broken or unreachable. <a href="https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/data/resource_source_audit.csv">View the results</a> or <a href="https://github.com/ChaoYue0307/awesome-loop-engineering/issues/new?template=annotation-correction.yml">report a correction</a>.</p>
+          <p class="audit-note"><strong>Last source check, 2026-07-21 UTC:</strong> 618 public links opened successfully, 2 required access, 49 point to files in this repository, and none were broken or unreachable. <a href="https://github.com/ChaoYue0307/awesome-loop-engineering/blob/main/data/resource_source_audit.csv">View the results</a> or <a href="https://github.com/ChaoYue0307/awesome-loop-engineering/issues/new?template=annotation-correction.yml">report a correction</a>.</p>
           <div class="loop-layer-map" aria-label="Where different loop concepts live">
             <div class="loop-layer-item"><span>Inner loop</span><strong>Model recurrence</strong><p>A learned block refines hidden or latent state inside one inference. Included as an adjacent foundation.</p></div>
             <div class="loop-layer-item"><span>Task loop</span><strong>Agent execution</strong><p>The model reasons, calls tools, observes results, and updates working state while solving one task.</p></div>
@@ -1879,7 +1879,7 @@
               <button class="resource-collection" type="button" data-collection="Apply" aria-pressed="false"><i class="icon" data-lucide="package-open" aria-hidden="true"></i>Apply</button>
             </div>
             <div class="resource-summary">
-              <p class="resource-count" id="resource-count" aria-live="polite">Preparing 669 resources...</p>
+              <p class="resource-count" id="resource-count" aria-live="polite">Preparing 671 resources...</p>
               <label>
                 <span class="sr-only">Sort resources</span>
                 <select class="resource-select resource-sort" id="resource-sort">
@@ -2483,7 +2483,7 @@
           loading = true;
           explorer.dataset.indexState = 'loading';
           list.setAttribute('aria-busy', 'true');
-          count.textContent = 'Loading 669 resources...';
+          count.textContent = 'Loading 671 resources...';
           list.replaceChildren(createText('p', 'resource-loading', 'Loading resources...'));
 
           fetch('assets/resources.json')

docs/x-v10-671.html

@@ -0,0 +1,44 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <title>Awesome Loop Engineering: 671 resources from model recurrence to agent operations</title>
+    <meta name="description" content="Explore 671 resources, then build recurring AI-agent systems with 22 patterns, 22 contracts, and 8 runtime starters.">
+    <meta name="robots" content="noindex,follow">
+    <link rel="canonical" href="https://chaoyue0307.github.io/awesome-loop-engineering/x-v10-671.html">
+    <meta property="og:type" content="website">
+    <meta property="og:title" content="Awesome Loop Engineering">
+    <meta property="og:description" content="669 papers, docs, tools, benchmarks, and guides, plus 22 patterns, 22 contracts, and 8 starters for recurring AI-agent systems.">
+    <meta property="og:url" content="https://chaoyue0307.github.io/awesome-loop-engineering/x-v10-671.html">
+    <meta property="og:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-671-v10.png">
+    <meta property="og:image:width" content="1280">
+    <meta property="og:image:height" content="640">
+    <meta property="og:image:alt" content="Awesome Loop Engineering: 671 resources, 22 patterns, 22 contracts, and 8 runtime starters.">
+    <meta name="twitter:card" content="summary_large_image">
+    <meta name="twitter:title" content="Awesome Loop Engineering">
+    <meta name="twitter:description" content="From looped models to verified agent work and bounded operations.">
+    <meta name="twitter:image" content="https://chaoyue0307.github.io/awesome-loop-engineering/assets/social-preview-671-v10.png">
+    <meta name="twitter:image:alt" content="Awesome Loop Engineering: design loops, not just prompts.">
+    <link rel="icon" href="assets/favicon-32.png" type="image/png" sizes="32x32">
+    <style>
+      :root { color-scheme: light; font-family: Inter, ui-sans-serif, system-ui, sans-serif; }
+      body { display: grid; min-height: 100vh; margin: 0; place-items: center; background: #f7f9fc; color: #111827; }
+      main { width: min(88vw, 420px); text-align: center; }
+      img { width: 72px; height: 72px; }
+      p { color: #526176; line-height: 1.6; }
+      a { color: #1d4ed8; font-weight: 700; }
+    </style>
+    <script>
+      window.location.replace("https://chaoyue0307.github.io/awesome-loop-engineering/?utm_source=x&utm_medium=social&utm_campaign=v0.10.0");
+    </script>
+  </head>
+  <body>
+    <main>
+      <img src="assets/awesome-loop-engineering-logo.png" alt="Awesome Loop Engineering">
+      <h1>Awesome Loop Engineering</h1>
+      <p>Opening the interactive Resource Atlas.</p>
+      <p><a href="https://chaoyue0307.github.io/awesome-loop-engineering/?utm_source=x&amp;utm_medium=social&amp;utm_campaign=v0.10.0">Continue to the site</a></p>
+    </main>
+  </body>
+</html>

examples/agent-memory-lifecycle-loop.json

@@ -0,0 +1,103 @@
+{
+  "name": "Agent memory lifecycle loop",
+  "objective": "Carry governed operational memory between recurring agent runs: explicitly written, scoped on recall, audited, consolidated, and deletable.",
+  "trigger": {
+    "type": "scheduled",
+    "cadence_or_event": "Weekly consolidation pass, plus a session-end candidate-write hook after each run."
+  },
+  "intake": {
+    "sources": [
+      "candidate records from recent runs",
+      "live memory store with provenance",
+      "retention and privacy policy",
+      "audit log",
+      "recall-scope definitions"
+    ],
+    "selection_rule": "Process only candidate records written since the last pass and live records due for audit, expiry, or review."
+  },
+  "workspace": {
+    "isolation": "Staging area for candidate records; the live store changes only through gated promotion.",
+    "allowed_actions": [
+      "distill run artifacts into candidate records",
+      "deduplicate and merge with provenance",
+      "audit for staleness, contradiction, and scope creep",
+      "apply logged expiries and redactions",
+      "run recall-scope test queries"
+    ],
+    "disallowed_actions": [
+      "store raw transcripts or credentials",
+      "delete audited records silently",
+      "write to the live store without promotion",
+      "broaden a record's recall scope without review"
+    ]
+  },
+  "context": {
+    "required_files": [
+      "memory store manifest",
+      "retention and privacy policy",
+      "recall-scope map",
+      "audit log"
+    ],
+    "runtime_sources": [
+      "session-end run artifacts",
+      "quarantine queue",
+      "recall test results"
+    ]
+  },
+  "agents": [
+    {
+      "role": "recorder",
+      "responsibility": "Distill each finished run into explicit candidate records instead of raw transcripts."
+    },
+    {
+      "role": "auditor",
+      "responsibility": "Flag stale, contradictory, sensitive, or unattributed records and quarantine failures."
+    },
+    {
+      "role": "consolidator",
+      "responsibility": "Merge duplicates, promote repeated lessons to principle-level records, and retire superseded ones."
+    },
+    {
+      "role": "recall gate",
+      "responsibility": "Scope what each future run may read by task and access class, and verify scopes with test recalls."
+    }
+  ],
+  "verification": {
+    "gates": [
+      "every record carries provenance, scope, and an expiry or review date",
+      "no raw transcript, credential, or policy-barred content in the live store",
+      "contradictory records cannot coexist unresolved",
+      "recall tests retrieve intended records and nothing out of scope",
+      "deletions and redactions are logged and reversible for the review window"
+    ],
+    "receipts": [
+      "consolidation receipt linking merged, promoted, and retired records",
+      "audit log entries for every write, merge, prune, and redaction"
+    ]
+  },
+  "state": {
+    "artifacts": [
+      "memory store with typed records",
+      "audit log of writes, merges, prunes, and redactions",
+      "consolidation receipts"
+    ],
+    "update_rule": "Only promoted records enter the live store; quarantined and expired records never recall; every mutation appends to the audit log."
+  },
+  "budget": {
+    "max_retries": 1,
+    "max_runtime_minutes": 60
+  },
+  "escalation": {
+    "conditions": [
+      "sensitive or ambiguous records",
+      "contradiction between operator instruction and remembered preference",
+      "retention-policy questions",
+      "records whose provenance does not trace to a legitimate run"
+    ],
+    "destination": "Review queue with the quarantined records and their provenance evidence."
+  },
+  "exit": {
+    "success": "All candidates processed, audits clean, recall-scope tests pass, and the receipt is written.",
+    "stop_without_success": "Churn cap reached or policy failures found; the live store keeps its last audited state."
+  }
+}

examples/fleet-coordination-loop.json

@@ -0,0 +1,102 @@
+{
+  "name": "Fleet coordination loop",
+  "objective": "Land many parallel agent-authored changes on one codebase without collisions, duplicate work, or unverified merges.",
+  "trigger": {
+    "type": "event",
+    "cadence_or_event": "Continuous while the task backlog is non-empty, within a wall-clock window."
+  },
+  "intake": {
+    "sources": [
+      "task backlog with priorities and dependencies",
+      "claim ledger",
+      "repository protected-branch and required-check configuration",
+      "fleet policy for concurrency and budgets"
+    ],
+    "selection_rule": "Dispatch only unclaimed ready tasks whose dependencies have landed, up to the concurrency cap."
+  },
+  "workspace": {
+    "isolation": "One worktree, branch, or container per worker; no shared mutable checkout.",
+    "allowed_actions": [
+      "claim a task through the ledger",
+      "implement the task in the owned workspace",
+      "submit a candidate branch",
+      "run the landing gate",
+      "land serially through the merge queue"
+    ],
+    "disallowed_actions": [
+      "push directly to the integration branch",
+      "work an unclaimed task",
+      "edit files claimed by another live task",
+      "land a candidate that has not passed the gate after final rebase"
+    ]
+  },
+  "context": {
+    "required_files": [
+      "task backlog",
+      "claim ledger",
+      "landing-gate definition",
+      "merge-queue state"
+    ],
+    "runtime_sources": [
+      "live claim ledger",
+      "merge-queue state",
+      "landing-gate results"
+    ]
+  },
+  "agents": [
+    {
+      "role": "dispatcher",
+      "responsibility": "Claim and assign tasks, reconcile stale claims, and enforce the concurrency cap."
+    },
+    {
+      "role": "worker",
+      "responsibility": "Implement one bounded task in an isolated workspace and submit a candidate change."
+    },
+    {
+      "role": "verifier",
+      "responsibility": "Run the landing gate on every candidate independently of the worker that produced it."
+    },
+    {
+      "role": "integrator",
+      "responsibility": "Serialize landings through the merge queue, rebasing and re-verifying on conflict."
+    }
+  ],
+  "verification": {
+    "gates": [
+      "full landing gate passes after the candidate's final rebase",
+      "diff stays inside the claimed task scope",
+      "no two live claims cover the same task or protected files",
+      "integration branch stays green; a red landing halts the queue",
+      "every landing has a receipt linking task, worker, gate results, and commit"
+    ],
+    "receipts": [
+      "landing receipt linking task, worker, gate results, and commit",
+      "fleet telemetry snapshot per wave"
+    ]
+  },
+  "state": {
+    "artifacts": [
+      "claim ledger",
+      "merge-queue state and landing receipts",
+      "fleet telemetry"
+    ],
+    "update_rule": "Claims and landings mutate only through the ledger and merge queue; unfinished tasks return to the backlog with their evidence attached."
+  },
+  "budget": {
+    "max_retries": 2,
+    "max_runtime_minutes": 480
+  },
+  "escalation": {
+    "conditions": [
+      "a task fails its gate repeatedly",
+      "two tasks genuinely require the same files",
+      "gate infrastructure breaks",
+      "landing order forces a product decision"
+    ],
+    "destination": "Escalation queue with the candidate branch, gate evidence, and claim history."
+  },
+  "exit": {
+    "success": "Backlog empty with all landings receipted and the integration branch green.",
+    "stop_without_success": "Budget spent or the branch cannot be kept green; remaining tasks return to the backlog with evidence."
+  }
+}

meta/DISTRIBUTION.md

@@ -8,7 +8,7 @@ Status as of 2026-07-20:
 
 | Surface | Status | Evidence / next action |
 | --- | --- | --- |
-| GitHub release | Complete | [`v0.9.0`](https://github.com/ChaoYue0307/awesome-loop-engineering/releases/tag/v0.9.0) packages 669 resources, 20 patterns and contracts, 8 runtime starters, and explicit model-to-operations scope facets. |
+| GitHub release | Complete | [`v0.9.0`](https://github.com/ChaoYue0307/awesome-loop-engineering/releases/tag/v0.9.0) packages 671 resources, 22 patterns and contracts, 8 runtime starters, and explicit model-to-operations scope facets. |
 | Release announcement | Ready | Publish the v0.9.0 summary from [`posts/launch.md`](../posts/launch.md); [Discussion #9](https://github.com/ChaoYue0307/awesome-loop-engineering/discussions/9) is the low-noise release and contributor feed. |
 | GitHub Explore topic | Complete | [`github/explore#5209`](https://github.com/github/explore/pull/5209) merged the `loop-engineering` topic page. |
 | Canonical Awesome directory | Policy objection | [`sindresorhus/awesome#4339`](https://github.com/sindresorhus/awesome/pull/4339) is open and lint passes, but the owner stated that AI-generated lists are not accepted. Do not treat further visual polish as a path around that policy or misrepresent automated assistance. |
@@ -50,7 +50,7 @@ Use GitHub-native discovery in this order.
   ```
 
 - **Pin the repository** on your profile, and add it to your profile README (profile-level promotion).
-- **About description.** Keep this concise value statement aligned with the README and project site: `🔁 Build reliable recurring AI-agent systems: 669 resources, 20 operational patterns, 20 loop contracts, 8 runtime starters, an interactive atlas, and a structured dataset.`
+- **About description.** Keep this concise value statement aligned with the README and project site: `🔁 Build reliable recurring AI-agent systems: 671 resources, 22 operational patterns, 22 loop contracts, 8 runtime starters, an interactive atlas, and a structured dataset.`
 - **Custom social preview** (repo Settings, Social preview) so shared links lead with the same four durable proof points. This is UI-only; it cannot be set through the API.
 - **Releases and Discussions** provide a lower-noise subscription path. Recommend `Watch -> Custom -> Releases and Discussions`; do not ask readers to watch every repository event.
 
@@ -95,7 +95,7 @@ Use the [English launch article](../posts/launch.md) or [Chinese launch article]
 ```text
 🔁 Awesome Loop Engineering maps how recurring AI agents should act, verify, remember, retry, and stop.
 
-669 resources · 20 patterns · 20 contracts · 8 starters
+671 resources · 22 patterns · 22 contracts · 8 starters
 
 Explore and build:
 https://github.com/ChaoYue0307/awesome-loop-engineering
@@ -105,7 +105,7 @@ https://github.com/ChaoYue0307/awesome-loop-engineering
 
 Attach [`assets/social-preview.png`](../assets/social-preview.png) directly when the platform permits it. The post is 264 characters before platform-specific URL shortening.
 
-For a fresh link card on X or LinkedIn, share <https://chaoyue0307.github.io/awesome-loop-engineering/x-v10-669.html>. The versioned URL points to the current preview image, records the `v0.10.0` campaign, and redirects readers to the interactive site.
+For a fresh link card on X or LinkedIn, share <https://chaoyue0307.github.io/awesome-loop-engineering/x-v10-671.html>. The versioned URL points to the current preview image, records the `v0.10.0` campaign, and redirects readers to the interactive site.
 
 ### Chinese Short Post
 
@@ -132,7 +132,7 @@ The release includes:
 - 669 papers, docs, tools, benchmarks, and guides
 - 29 model-layer resources, including Loopie, labeled as adjacent foundations
 - an interactive Resource Atlas
-- 20 operational patterns and 20 adaptable loop contracts
+- 22 operational patterns and 22 adaptable loop contracts
 - 8 runtime starters: 3 executables and 5 copy/paste templates
 - 50-field CSV, JSONL, and Parquet exports mirrored to Hugging Face
 - source and evidence fields, with point-in-time link-check results
@@ -146,11 +146,11 @@ https://github.com/ChaoYue0307/awesome-loop-engineering
 Use a regular submission, not Show HN.
 
 ```text
-Title: Awesome Loop Engineering: 669 resources from looped models to agent operations
+Title: Awesome Loop Engineering: 671 resources from looped models to agent operations
 
 I have been mapping recurring AI-agent systems: how work enters, agents act, evidence gates results, state survives, and the system retries, escalates, or exits.
 
-The repository connects 669 papers, docs, tools, benchmarks, and guides across model, agent, harness, workflow, operations, and evaluation layers with 20 operational patterns, 20 adaptable contracts, 8 runtime starters, an interactive Resource Atlas, and a structured Hugging Face dataset. Model-level recurrence is labeled as an adjacent foundation rather than a complete operational loop. Corrections to source links, taxonomy, and summaries are especially welcome.
+The repository connects 669 papers, docs, tools, benchmarks, and guides across model, agent, harness, workflow, operations, and evaluation layers with 22 operational patterns, 22 adaptable contracts, 8 runtime starters, an interactive Resource Atlas, and a structured Hugging Face dataset. Model-level recurrence is labeled as an adjacent foundation rather than a complete operational loop. Corrections to source links, taxonomy, and summaries are especially welcome.
 
 https://github.com/ChaoYue0307/awesome-loop-engineering
 ```

meta/social-preview.html

@@ -79,9 +79,9 @@
         <p>Model / agent / operations</p>
       </div>
       <div class="stats">
-        <div class="stat"><b>669</b><span>Papers, docs, tools + guides</span></div>
-        <div class="stat"><b>20</b><span>Operational patterns</span></div>
-        <div class="stat"><b>20</b><span>Schema-checked contracts</span></div>
+        <div class="stat"><b>671</b><span>Papers, docs, tools + guides</span></div>
+        <div class="stat"><b>22</b><span>Operational patterns</span></div>
+        <div class="stat"><b>22</b><span>Schema-checked contracts</span></div>
         <div class="stat"><b>8</b><span>Runtime starters</span></div>
       </div>
     </main>

patterns/MATRIX.md

@@ -1,6 +1,6 @@
 # Pattern Matrix
 
-A side-by-side comparison of all 20 loops in the [pattern library](README.md). Use it to pick a pattern, compare trade-offs, or check that your own loop names the same parts.
+A side-by-side comparison of all 22 loops in the [pattern library](README.md). Use it to pick a pattern, compare trade-offs, or check that your own loop names the same parts.
 
 ## How To Read This Matrix
 
@@ -37,6 +37,8 @@ Every pattern links to its full write-up and a schema-valid contract in [`exampl
 | [Incident response loop](incident-response-loop.md) | An incident just paged | Event: alert; refresh 2-5 min | Alert, dashboards, traces, changes | Incident timeline, hypotheses | Severity backed by impact evidence | 2 retries / incident life | Outage, data loss, or mitigation needed | Custom durable runtime, shell/cron |
 | [Data-quality loop](data-quality-loop.md) | A dataset keeps drifting | Event: refresh; or nightly | New version, schema, prior baseline | Profile metrics, accepted exceptions | Hard rules pass before promotion | 2 retries / 90 min | Schema change or regulated fields | Custom durable runtime, shell/cron |
 | [Knowledge freshness loop](knowledge-freshness-loop.md) | An agent knowledge index is stale | Daily refresh + source change | Approved sources, manifest, retrieval evals | Versioned index, provenance, tombstones | Freshness, retrieval, leakage, atomic promotion | 2 retries / 120 min | Source conflict, access change, sensitive data | Custom durable runtime, scheduled task |
+| [Agent memory lifecycle loop](agent-memory-lifecycle-loop.md) | Recurring runs cold-start or repeat mistakes | Weekly consolidation + session end | Candidate records, live store, retention policy | Typed records, audit log, receipts | Provenance, scope, contradiction, recall tests | 1 retry / 60 min | Sensitive records, contradictions, injection artifacts | Scheduled task, session hook |
+| [Fleet coordination loop](fleet-coordination-loop.md) | Parallel agents collide or duplicate work | Continuous while backlog non-empty | Task backlog, claim ledger, fleet policy | Claim ledger, merge queue, telemetry | Landing gate after rebase, scoped diff, green branch | 2 retries / 480 min | Repeated gate failures, entangled tasks, product decisions | Merge queue, worktrees, CI |
 | [Release-note loop](release-note-loop.md) | Release notes are a chore | Event: release branch or tag | Merged PRs, issues, commits | Processed IDs, draft changelog | Every entry links to a source | 2 retries / 60 min | Security advisories or wording calls | gh-aw, Codex automations |
 | [Model-routing loop](model-routing-loop.md) | Model choice is ad hoc | Schedule + model/price change | Per-task telemetry, eval results | Task-class metrics, routing log | Replays hold quality within tolerance | 2 retries / 120 min | Privacy tier or quality-cost tradeoffs | Custom durable runtime |
 

patterns/agent-memory-lifecycle-loop.md

@@ -0,0 +1,106 @@
+# Agent Memory Lifecycle Loop
+
+## Objective
+
+Keep recurring agent work from cold-starting every run while the memory that carries between runs stays governed: explicitly written, scoped on recall, audited, consolidated, and deletable.
+
+## Use This When
+
+- An agent runs repeatedly against the same project, product, or user and repeats mistakes it already made.
+- Prior-session state materially changes what the next run should do.
+- Unbounded transcript carryover has caused stale, contradictory, or sensitive context to leak into later runs.
+
+Use the knowledge-freshness loop when the durable product is a retrieval corpus assembled from external sources. Use this pattern when the durable product is the agent's own operational memory: lessons, decisions, preferences, warnings, and open threads.
+
+## Trigger
+
+- Schedule: consolidation pass daily or weekly; audit pass on a longer cadence.
+- Event: session end, task completion, memory-store size or age threshold, or a policy change that affects retention.
+- Manual bootstrap: "consolidate and audit the agent memory for <project>."
+
+## Intake
+
+- Candidate writes from recent runs: lessons, decisions, preferences, failures worth remembering.
+- The current memory store with per-record provenance, timestamps, and access class.
+- Retention, privacy, and escalation policy for what may be remembered at all.
+
+## Agents
+
+- Recorder: distills a finished run into explicit candidate records, never raw transcripts.
+- Auditor: flags stale, contradictory, sensitive, or unattributed records.
+- Consolidator: merges duplicates, promotes repeated lessons to principles, and prunes superseded records.
+- Recall gate: scopes what each future run may read, by task and access class.
+
+## Workspace And Permissions
+
+- Memory writes go to a staging area; promotion to the live store is a separate gated step.
+- Allow reads of run artifacts and the live store; allow writes only to staging and the audit log.
+- Disallow storing raw transcripts, credentials, or personal data barred by policy, and disallow silent deletion of audited records.
+
+## Durable State
+
+- The memory store itself: typed records with provenance, scope, timestamps, and expiry or review dates.
+- Audit log of writes, merges, prunes, redactions, and the run that caused each.
+- Consolidation receipts: what was merged or promoted, and which records were retired.
+
+## Loop Steps
+
+1. Collect candidate records from runs since the last pass.
+1. Reject candidates that violate retention or privacy policy at write time.
+1. Deduplicate against the live store; merge near-duplicates with provenance preserved.
+1. Audit the live store for staleness, contradiction, and scope creep; quarantine failures.
+1. Consolidate repeated instance-level lessons into principle-level records.
+1. Apply expiries: delete, redact, or send to review anything past its window.
+1. Write the receipt and update recall scopes for the next run.
+
+## Verification Gates
+
+- Every record has provenance, a scope, and an expiry or review date.
+- No raw transcript, credential, or policy-barred content is present in the live store.
+- Contradictory records cannot coexist unresolved; the newer record must cite what it supersedes.
+- Recall tests confirm the next run retrieves the intended records and nothing out of scope.
+- Deletions and redactions are logged and reversible for the review window.
+
+## Budget And Exit
+
+- Max runtime: 30-60 minutes per consolidation pass.
+- Max churn: cap the fraction of the store that one pass may merge or delete.
+- Stop when candidates are exhausted, the churn cap is reached, or the audit finds policy failures that need a human.
+
+## Escalation
+
+Escalate for sensitive or ambiguous records, contradictions between an operator instruction and remembered preference, retention-policy questions, and any suspected memory-injection artifact: records whose provenance does not trace to a legitimate run.
+
+## Loop Instruction
+
+```text
+Consolidate agent memory for <project>.
+Distill recent runs into explicit candidate records; never store raw transcripts.
+Deduplicate, audit for staleness and contradiction, consolidate repeated lessons into principles,
+and apply expiries with logged, reversible deletions.
+Verify recall scope with test queries before finishing.
+Escalate sensitive records, unresolved contradictions, and any record whose provenance
+does not trace to a legitimate run.
+```
+
+## Worked Example
+
+A team runs a nightly PR-babysitter loop over one repository. Its weekly memory pass distills the week's runs into candidate lessons, merges three near-duplicate records about a flaky integration test into one principle with provenance, expires a remembered workaround that a merged fix made obsolete, quarantines one record whose provenance traces to no known run, and verifies with test recalls that the next babysitter run retrieves the flaky-test principle but not the retired workaround.
+
+## Failure Modes
+
+- Remembering transcripts instead of distilled records, so recall floods future runs with stale context.
+- Treating the most recent or most assertive record as true when records contradict.
+- Memory that only grows: no expiry, no consolidation, no deletion path.
+- Poisoned records entering the store because writes are not gated by provenance checks.
+- Recall without scoping, so unrelated tasks read each other's remembered state.
+
+## Example Contract
+
+- [`examples/agent-memory-lifecycle-loop.json`](../examples/agent-memory-lifecycle-loop.json)
+
+## References
+
+- [Always-On Agents: A Survey of Persistent Memory, State, and Governance in LLM Agents](https://arxiv.org/abs/2606.30306) - Finds the literature over-indexes on accumulating state and under-indexes on governing, recovering, or deleting it.
+- [The Compliance Trap: Diagnosing How AI Agents Consume Conflicting Memory](https://arxiv.org/abs/2607.10608) - Shows agents tend to comply with the most recent or most assertive memory record rather than the correct one.
+- [When Claws Remember but Do Not Tell: Stealthy Memory Injection in Persistent Personal Agents](https://arxiv.org/abs/2607.05189) - Demonstrates why memory writes need provenance gates: one poisoned input can silently steer future unattended runs.

patterns/fleet-coordination-loop.md

@@ -0,0 +1,104 @@
+# Fleet Coordination Loop
+
+## Objective
+
+Run many coding agents in parallel against one codebase or backlog without merge collisions, duplicated work, or unreviewed changes reaching the integration branch.
+
+## Use This When
+
+- More than a handful of agents work the same repository or task queue concurrently.
+- Agent throughput is limited by coordination, not generation: collisions, rework, and review pile-ups.
+- You can define a deterministic landing gate that every parallel change must pass before integration.
+
+Use the PR-babysitter loop to shepherd a single change. Use this pattern when the unit of work is the fleet: assigning, isolating, verifying, and landing many parallel changes.
+
+## Trigger
+
+- Schedule: continuous while the backlog is non-empty, or in timed waves.
+- Event: backlog growth past a threshold, a milestone branch opening, or a batch of verified tasks becoming ready.
+- Manual bootstrap: "work this backlog with <N> parallel agents until the queue is empty or budget is spent."
+
+## Intake
+
+- The task backlog with priorities, dependencies, and claim status.
+- Repository state: protected branches, required checks, and the landing gate definition.
+- Fleet policy: max concurrency, per-agent budgets, and which task classes may run unattended.
+
+## Agents
+
+- Dispatcher: claims tasks, assigns them to workers, and prevents duplicate claims.
+- Workers: implement one bounded task each inside an isolated workspace.
+- Verifier: runs the landing gate on each candidate change independently of the worker that produced it.
+- Integrator: serializes landings through a merge queue and resolves ordering.
+
+## Workspace And Permissions
+
+- One isolated worktree, branch, or container per worker; no shared mutable checkout.
+- Allow workers to read the repository and write only to their own workspace and claim record.
+- Disallow direct pushes to the integration branch, cross-worker file claims, and landing without the gate.
+
+## Durable State
+
+- The claim ledger: which agent owns which task, started when, with what budget remaining.
+- Merge-queue state: ordered candidates, gate results, and landing receipts.
+- Fleet telemetry: per-task cost, retry counts, collision events, and gate failures for tuning concurrency.
+
+## Loop Steps
+
+1. Refresh the backlog and reconcile stale claims from dead or expired workers.
+1. Dispatch unclaimed ready tasks to idle workers up to the concurrency cap.
+1. Each worker implements its task in isolation and submits a candidate change.
+1. Verify each candidate with the landing gate: build, tests, lint, and scoped-diff checks.
+1. Queue passing candidates; land them serially, rebasing or re-verifying on conflict.
+1. Return failed candidates to their worker with evidence, up to the retry budget.
+1. Record receipts and telemetry; adjust concurrency if collisions or costs trend up.
+
+## Verification Gates
+
+- Every landed change passed the full landing gate after its final rebase, not before.
+- Each change's diff stays inside its claimed task scope.
+- No two live claims cover the same task or the same protected files.
+- The integration branch stays green: a failed landing halts the queue, not just the one change.
+- Every landing has a receipt linking task, worker, gate results, and commit.
+
+## Budget And Exit
+
+- Per-task: bounded retries and runtime before the task returns to the backlog for a human.
+- Per-fleet: max concurrent workers, total token or cost ceiling, and a wall-clock window.
+- Stop when the backlog is empty, the budget is spent, or the integration branch cannot be kept green.
+
+## Escalation
+
+Escalate when the same task fails its gate repeatedly, two tasks genuinely require the same files, gate infrastructure itself breaks, or landing order forces a product decision no worker owns.
+
+## Loop Instruction
+
+```text
+Work the backlog at <repo> with up to <N> parallel workers.
+One isolated workspace per worker; claim tasks through the ledger, never work an unclaimed task.
+Every candidate change must pass the landing gate after final rebase; land serially through the queue.
+Return gate failures to the owning worker with evidence, within retry budget.
+Halt the queue and escalate if the integration branch goes red or claims conflict.
+```
+
+## Worked Example
+
+A migration backlog of 140 mechanical refactor tasks runs with twelve workers. The dispatcher hands out tasks by dependency order; each worker edits in its own worktree and submits a branch. The verifier runs build, tests, and a scoped-diff check per candidate; the merge queue lands nine changes per hour serially, bounces two candidates back with failing-test evidence, and reconciles one stale claim after a worker times out. The fleet exits with 137 landed, 3 escalated as genuinely entangled, and per-task receipts for review.
+
+## Failure Modes
+
+- Shared checkouts, so parallel edits trample each other before verification.
+- Verifying before the final rebase and landing changes that break the branch anyway.
+- Two workers silently duplicating one task because claims are advisory.
+- Per-agent monitors missing fleet-level effects: correlated failures and coordinated drift.
+- Scaling concurrency past what the merge queue and reviewers can absorb, converting throughput into backlog.
+
+## Example Contract
+
+- [`examples/fleet-coordination-loop.json`](../examples/fleet-coordination-loop.json)
+
+## References
+
+- [Scaling Long-Running Autonomous Coding](https://cursor.com/blog/scaling-agents) - Cursor's coordination designs, from flat agents with locking to a planner, worker, and judge hierarchy pushing to a single branch.
+- [Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors](https://arxiv.org/abs/2607.07368) - Shows per-agent monitors weaken as more agents coordinate, motivating fleet-level gates and telemetry.
+- [claude-code-merge-queue](https://github.com/funador/claude-code-merge-queue) - A minimal landing gate: serialize parallel agents through a FIFO queue that blocks integration until a check command passes.

posts/launch.md

@@ -2,14 +2,14 @@
 
 Awesome Loop Engineering v0.9.0 maps recurrence from a model's inner computation to an agent's tools and the outer operating loop that verifies, remembers, retries, escalates, and stops.
 
-Explore 669 resources without collapsing looped models, agent execution, harnesses, workflows, and production operations into one idea.
+Explore 671 resources without collapsing looped models, agent execution, harnesses, workflows, and production operations into one idea.
 
 ## What You Can Use
 
 - 669 papers, official docs, tools, benchmarks, and guides linked to the original work
 - 29 model-layer resources, from Universal Transformers and Huginn to Loopie, LoopCoder, and LoopWM, labeled as adjacent foundations rather than complete agent loops
-- 20 operational patterns organized by build, operate, optimize, and govern use cases
-- 20 schema-checked loop contracts, one for every pattern
+- 22 operational patterns organized by build, operate, optimize, and govern use cases
+- 22 schema-checked loop contracts, one for every pattern
 - 8 runtime starters: 3 dependency-light executables and 5 copy/paste runtime templates
 - an interactive Resource Atlas for filtering by goal, loop layer, lifecycle stage, artifact type, and evidence class
 - 50-field CSV, JSONL, and Parquet exports mirrored as a Hugging Face dataset

posts/launch.zh-CN.md

@@ -2,11 +2,11 @@
 
 Awesome Loop Engineering v0.9.0 建立了一张从模型内部递归、Agent 执行、Harness 到外层运行系统的完整地图。
 
-可浏览 669 篇论文、官方文档、工具、benchmark 与实践指南，同时保持清晰边界：模型内的循环提供自适应计算，但不等于具备外部验证、持久状态、预算、人类交接与退出规则的运行级 Loop。
+可浏览 671 篇论文、官方文档、工具、benchmark 与实践指南，同时保持清晰边界：模型内的循环提供自适应计算，但不等于具备外部验证、持久状态、预算、人类交接与退出规则的运行级 Loop。
 
 ## 可直接使用
 
-- 669 篇论文、官方文档、工具、benchmark 与实践指南，并直接链接原始来源
+- 671 篇论文、官方文档、工具、benchmark 与实践指南，并直接链接原始来源
 - 29 个模型层资源，覆盖 Universal Transformers、Huginn、Loopie、LoopCoder、LoopWM 等方向
 - 20 个按 build、operate、optimize、govern 组织的 operational patterns
 - 20 个经过 schema 校验的 loop contracts，每个 pattern 对应一个