@piagent/platform 1.2.17 → 1.3.0-ie.9

package/docs-site/architecture.html

@@ -30,7 +30,7 @@
         <span>Pi Agent Platform</span>
       </a>
       <div class="top-actions">
-        <span class="pill"><span class="pill-dot"></span>v1.2.17 docs · VI</span>
+        <span class="pill"><span class="pill-dot"></span>v1.3.0-ie.9 docs · VI</span>
         <nav class="language-switch" aria-label="Ngôn ngữ">
           <a class="language-option current" href="/architecture" hreflang="vi" lang="vi" aria-current="true">VI</a>
           <a class="language-option" href="/en/architecture" hreflang="en" lang="en">EN</a>
@@ -60,6 +60,7 @@
             <p class="nav-title">Get started</p>
             <a class="nav-link" href="/">Overview</a>
             <a class="nav-link" href="/quickstart">Quickstart</a>
+            <a class="nav-link" href="/whats-new">Có gì mới</a>
             <a class="nav-link" href="/scope">Phạm vi platform</a>
           </div>
           <div class="nav-group">
@@ -134,6 +135,40 @@
   </div>
 </section>
 
+<section class="section">
+  <div class="section-head">
+    <p class="eyebrow">Adaptive runtime</p>
+    <h2 id="adaptive-runtime">Piagent tách runtime decision khỏi Pi wiring</h2>
+    <p>Harness giữ nguyên model người dùng chọn, nhưng mỗi task có context plan nhỏ hơn, có citation, có task journal bền vững và recovery decision rõ ràng.</p>
+  </div>
+  <div class="grid cols-3">
+    <article class="card">
+      <h3>Context planner</h3>
+      <p><span class="inline-code">runtime/context/adaptive-planner.ts</span> chọn budget, retrieval lane, code inclusion và plan receipt theo phase, context pressure, model capability và thinking level.</p>
+    </article>
+    <article class="card">
+      <h3>Task journal</h3>
+      <p><span class="inline-code">extensions/task-journal.js</span> ghi checkpoint cho start, progress, mutation, verifier và completion. Resume tin task contract trước; retention compaction giữ hash anchor.</p>
+    </article>
+    <article class="card">
+      <h3>Repository memory</h3>
+      <p><span class="inline-code">extensions/repository-memory.js</span> chỉ lưu fact có citation và expiry; từ chối raw prompt/output, OAuth material, API key và secret. Hint chỉ dùng token budget còn dư.</p>
+    </article>
+    <article class="card">
+      <h3>Model capability</h3>
+      <p><span class="inline-code">runtime/model/capabilities.ts</span> đọc model family, context window và thinking level làm tín hiệu budget, không đổi provider và không cần API key.</p>
+    </article>
+    <article class="card">
+      <h3>Verification intelligence</h3>
+      <p><span class="inline-code">extensions/verification-intelligence.js</span> phân loại lỗi và chọn verify group theo task scope; scope chạm cả frontend/backend dùng full source verifier.</p>
+    </article>
+    <article class="card">
+      <h3>Execution backend</h3>
+      <p><span class="inline-code">extensions/execution-backend.js</span> giữ Pi OAuth local làm owner; mutation bị block nếu experimental adapter được yêu cầu nhưng chưa có.</p>
+    </article>
+  </div>
+</section>
+
 <section class="section">
   <div class="section-head">
     <p class="eyebrow">Ownership</p>
@@ -146,6 +181,8 @@
       <tbody>
         <tr><td>Project profile + lock</td><td>Project</td><td><span class="inline-code">.pi/piagent-profile*.json</span></td><td>Có</td></tr>
         <tr><td>Task, trace, telemetry, capture</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/</span></td><td>Không</td></tr>
+        <tr><td>Task journal + checkpoint</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/task-journal/</span></td><td>Không</td></tr>
+        <tr><td>Repository memory fact</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/repository-memory/</span></td><td>Không</td></tr>
         <tr><td>Session history</td><td>Pi</td><td>Pi session store</td><td>Không</td></tr>
         <tr><td>Shared instructions</td><td>Project</td><td><span class="inline-code">AGENTS.md</span>, project docs</td><td>Có</td></tr>
       </tbody>

package/docs-site/benchmark.html

@@ -30,7 +30,7 @@
         <span>Pi Agent Platform</span>
       </a>
       <div class="top-actions">
-        <span class="pill"><span class="pill-dot"></span>v1.2.17 docs · VI</span>
+        <span class="pill"><span class="pill-dot"></span>v1.3.0-ie.9 docs · VI</span>
         <nav class="language-switch" aria-label="Ngôn ngữ">
           <a class="language-option current" href="/benchmark" hreflang="vi" lang="vi" aria-current="true">VI</a>
           <a class="language-option" href="/en/benchmark" hreflang="en" lang="en">EN</a>
@@ -60,6 +60,7 @@
             <p class="nav-title">Get started</p>
             <a class="nav-link" href="/">Overview</a>
             <a class="nav-link" href="/quickstart">Quickstart</a>
+            <a class="nav-link" href="/whats-new">Có gì mới</a>
             <a class="nav-link" href="/scope">Phạm vi platform</a>
           </div>
           <div class="nav-group">
@@ -104,7 +105,7 @@
 <section class="benchmark-summary" aria-labelledby="benchmark-verdict">
   <div class="benchmark-verdict">
     <div>
-      <p class="section-kicker">Production V1 · release evidence</p>
+      <p class="section-kicker">Production V1 · historical public regression evidence</p>
       <h2 id="benchmark-verdict">Piagent đạt 9,92/10</h2>
       <p>Hoàn thành toàn bộ 54 task, giữ Quality, Safety và Reliability ở 10/10, đồng thời dùng ít hơn 51,11% fresh token so với surface <code>codex-cli</code> trong phép đo ghép cặp.</p>
     </div>
@@ -285,6 +286,26 @@
   </div>
 </section>
 
+<section class="section">
+  <p class="section-kicker">Benchmark bands</p>
+  <h2 id="benchmark-bands">Chọn đúng suite theo quyết định cần đưa ra</h2>
+  <div class="table-wrap benchmark-table-wrap">
+    <table>
+      <thead><tr><th>Band</th><th>Khi nào dùng</th><th>Evidence tối thiểu</th><th>Claim được phép</th></tr></thead>
+      <tbody>
+        <tr><td>Core</td><td>Smoke nhanh trước khi làm release</td><td>4 scenario, 1 repeat, hidden verifier</td><td>Chỉ dùng để bắt regression</td></tr>
+        <tr><td>Production</td><td>Public regression cho release candidate</td><td>18 scenario, 3 repeat, <span class="inline-code">piagent</span> vs baseline</td><td>Claim chỉ giới hạn trong observed public suite; tách outcome coverage, comparable CI và failure-aware effort</td></tr>
+        <tr><td>Capability</td><td>Tìm trần năng lực của harness</td><td>4 bài lớn multi-file/multi-component với reference và mutation checks</td><td>Hill-climbing, không phải release hay generalization gate</td></tr>
+        <tr><td>Long-horizon</td><td>Thay đổi recovery/context và adopt repo lớn</td><td>Dedicated paid suite đang planned; hiện chạy Production cộng deterministic crash/replay tests</td><td>Chưa công bố điểm long-horizon cho tới khi suite được phát hành</td></tr>
+      </tbody>
+    </table>
+  </div>
+  <div class="callout">
+    <strong>Benchmark phải ghi đúng claim tier, không phải marketing copy.</strong>
+    <p>Một version không được claim tiết kiệm token nếu chưa pass quality/reliability, paired candidate-regression, outcome coverage, comparable-efficiency, failure-aware effort, scope/safety và protocol gate. So với raw Pi là causal harness ablation; so với <code>codex-cli</code> chỉ là external product reference.</p>
+  </div>
+</section>
+
 <section class="section">
   <p class="section-kicker">Provenance</p>
   <h2 id="minh-bach">Đọc kết quả đúng phạm vi</h2>
@@ -302,7 +323,7 @@
 
   <div class="callout">
     <strong>Giới hạn của kết luận</strong>
-    <p>Đây là synthetic production benchmark do maintainer xây dựng, không phải benchmark độc lập chứng minh Piagent tốt hơn trên mọi repository hoặc model. Raw report giữ private vì chứa seed tái tạo synthetic secret. Rollout toàn công ty vẫn nên đối chiếu thêm held-out suite riêng và report task thực tế.</p>
+    <p>Đây là public synthetic regression do maintainer xây dựng, không phải benchmark độc lập chứng minh Piagent tốt hơn trên mọi repository hoặc model. Generated values không tạo task family độc lập. Generalization cần external family-disjoint holdout, reference/mutation/human-calibration manifest, candidate freeze và shadow/canary task thực tế.</p>
   </div>
 
   <div class="command-grid">

package/docs-site/commands.html

@@ -4,7 +4,7 @@
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1" />
     <title>Command ngắn, rõ, ít token burn · Pi Agent Platform Docs</title>
-    <meta name="description" content="Ưu tiên dùng slash command vì mỗi command đã gói sẵn policy và flow. Thành viên không cần nhớ lại quy trình trong prompt." />
+    <meta name="description" content="Slash command gói sẵn policy và flow; một inspector menu duy nhất cho diff, command, safety và context. Thành viên không cần nhớ nhiều command rời." />
     <meta name="theme-color" content="#050505" />
     <link rel="canonical" href="https://piagent.io.vn/commands" />
     <link rel="alternate" hreflang="vi" href="https://piagent.io.vn/commands" />
@@ -12,7 +12,7 @@
     <link rel="alternate" hreflang="x-default" href="https://piagent.io.vn/commands" />
     <link rel="icon" href="/favicon.svg" type="image/svg+xml" />
     <meta property="og:title" content="Command ngắn, rõ, ít token burn · Pi Agent Platform Docs" />
-    <meta property="og:description" content="Ưu tiên dùng slash command vì mỗi command đã gói sẵn policy và flow. Thành viên không cần nhớ lại quy trình trong prompt." />
+    <meta property="og:description" content="Slash command gói sẵn policy và flow; một inspector menu duy nhất cho diff, command, safety và context. Thành viên không cần nhớ nhiều command rời." />
     <meta property="og:url" content="https://piagent.io.vn/commands" />
     <link rel="preconnect" href="https://fonts.googleapis.com" />
     <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
@@ -30,7 +30,7 @@
         <span>Pi Agent Platform</span>
       </a>
       <div class="top-actions">
-        <span class="pill"><span class="pill-dot"></span>v1.2.17 docs · VI</span>
+        <span class="pill"><span class="pill-dot"></span>v1.3.0-ie.9 docs · VI</span>
         <nav class="language-switch" aria-label="Ngôn ngữ">
           <a class="language-option current" href="/commands" hreflang="vi" lang="vi" aria-current="true">VI</a>
           <a class="language-option" href="/en/commands" hreflang="en" lang="en">EN</a>
@@ -60,6 +60,7 @@
             <p class="nav-title">Get started</p>
             <a class="nav-link" href="/">Overview</a>
             <a class="nav-link" href="/quickstart">Quickstart</a>
+            <a class="nav-link" href="/whats-new">Có gì mới</a>
             <a class="nav-link" href="/scope">Phạm vi platform</a>
           </div>
           <div class="nav-group">
@@ -99,7 +100,7 @@
             </nav>
             <div class="page-head">
               <h1>Command ngắn, rõ, ít token burn</h1>
-              <p>Ưu tiên dùng slash command vì mỗi command đã gói sẵn policy và flow. Thành viên không cần nhớ lại quy trình trong prompt.</p>
+              <p>Slash command gói sẵn policy và flow; một inspector menu duy nhất cho diff, command, safety và context. Thành viên không cần nhớ nhiều command rời.</p>
             </div>
 <section class="section">
 
@@ -111,8 +112,11 @@
         <span class="tag">/model</span>
         <span class="tag">/commands</span>
         <span class="tag">/piagent-status</span>
+        <span class="tag">/piagent-inspector</span>
         <span class="tag">/piagent-orchestration</span>
         <span class="tag">/usage</span>
+        <span class="tag">/usage efficiency</span>
+        <span class="tag">/task-preflight</span>
         <span class="tag">/context preflight</span>
         <span class="tag">/name</span>
         <span class="tag">/fresh</span>
@@ -185,6 +189,43 @@
     </article>
   </div>
 </section>
+
+<section class="section">
+  <p class="section-kicker">Product visibility</p>
+  <h2 id="product-status-command">Xem runtime đang quyết định gì</h2>
+  <p class="section-lead">Các command này đọc state deterministic, không cần model giải thích và không bắt buộc trong workflow hằng ngày.</p>
+  <div class="command-grid">
+    <article class="command-card">
+      <header><span>Preflight trước task</span><button class="clip-btn" data-clip="/task-preflight --json Review current repository state">Lấy lệnh</button></header>
+      <pre><code>/task-preflight --json Review current repository state</code></pre>
+      <p>Xem intent, risk, scope, runtime, solver route, phase, tool/helper recommendation, backend, approval và blocker trước khi bắt đầu.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Live task status</span><button class="clip-btn" data-clip="/piagent-status">Lấy lệnh</button></header>
+      <pre><code>/piagent-status</code></pre>
+      <p>Đọc Task Contract, checkpoint, verifier, recovery, helper, resume state và terminal receipt. Receipt chỉ approved khi hard completion gate trên working tree hiện tại pass.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Activity, diff và context</span><button class="clip-btn" data-clip="/piagent-inspector">Lấy lệnh</button></header>
+      <pre><code>/piagent-inspector</code></pre>
+      <p>Một menu duy nhất cho <span class="inline-code">summary</span>, <span class="inline-code">files</span>, <span class="inline-code">commands</span>, <span class="inline-code">security</span>, <span class="inline-code">context</span> và <span class="inline-code">toggle</span>. Panel bốn dòng tự hiện sát phía trên footer native; xanh lá là ổn, vàng cần chú ý, đỏ là failure/safety hoặc context từ 80%. Command chỉ đọc state local, không gọi model.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Efficiency có nguồn</span><button class="clip-btn" data-clip="/usage efficiency">Lấy lệnh</button></header>
+      <pre><code>/usage efficiency</code></pre>
+      <p>Xem context/tool utilization, verify, repair, retry và helper usage. Giá trị runtime không đo được giữ là <span class="inline-code">null</span>, không ước đoán.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Helper policy</span><button class="clip-btn" data-clip="/piagent-orchestration">Lấy lệnh</button></header>
+      <pre><code>/piagent-orchestration</code></pre>
+      <p>Xem solo-first policy, read-only roles, reviewer/Oracle recommendation, owned budget và one-writer invariant.</p>
+    </article>
+  </div>
+  <div class="callout">
+    <strong>Diff rõ nguồn, token không ước lượng.</strong>
+    <p>Footer status dùng file task đã quan sát để cập nhật nhanh; view <span class="inline-code">files</span> tính exact snapshot delta và đánh dấu <span class="inline-code">mixed-working-tree</span> nếu task đụng file đã dirty từ trước. Pi chỉ cung cấp model usage theo turn/session, nên token theo từng built-in tool luôn hiện unavailable.</p>
+  </div>
+</section>
             <nav class="pager" aria-label="Trang kế tiếp">
               <a class="pager-prev" href="/permissions">
                 <span class="pager-label">Trước</span>

package/docs-site/content/architecture.html

@@ -31,6 +31,40 @@
   </div>
 </section>
 
+<section class="section">
+  <div class="section-head">
+    <p class="eyebrow">Adaptive runtime</p>
+    <h2 id="adaptive-runtime">Piagent tách runtime decision khỏi Pi wiring</h2>
+    <p>Harness giữ nguyên model người dùng chọn, nhưng mỗi task có context plan nhỏ hơn, có citation, có task journal bền vững và recovery decision rõ ràng.</p>
+  </div>
+  <div class="grid cols-3">
+    <article class="card">
+      <h3>Context planner</h3>
+      <p><span class="inline-code">runtime/context/adaptive-planner.ts</span> chọn budget, retrieval lane, code inclusion và plan receipt theo phase, context pressure, model capability và thinking level.</p>
+    </article>
+    <article class="card">
+      <h3>Task journal</h3>
+      <p><span class="inline-code">extensions/task-journal.js</span> ghi checkpoint cho start, progress, mutation, verifier và completion. Resume tin task contract trước; retention compaction giữ hash anchor.</p>
+    </article>
+    <article class="card">
+      <h3>Repository memory</h3>
+      <p><span class="inline-code">extensions/repository-memory.js</span> chỉ lưu fact có citation và expiry; từ chối raw prompt/output, OAuth material, API key và secret. Hint chỉ dùng token budget còn dư.</p>
+    </article>
+    <article class="card">
+      <h3>Model capability</h3>
+      <p><span class="inline-code">runtime/model/capabilities.ts</span> đọc model family, context window và thinking level làm tín hiệu budget, không đổi provider và không cần API key.</p>
+    </article>
+    <article class="card">
+      <h3>Verification intelligence</h3>
+      <p><span class="inline-code">extensions/verification-intelligence.js</span> phân loại lỗi và chọn verify group theo task scope; scope chạm cả frontend/backend dùng full source verifier.</p>
+    </article>
+    <article class="card">
+      <h3>Execution backend</h3>
+      <p><span class="inline-code">extensions/execution-backend.js</span> giữ Pi OAuth local làm owner; mutation bị block nếu experimental adapter được yêu cầu nhưng chưa có.</p>
+    </article>
+  </div>
+</section>
+
 <section class="section">
   <div class="section-head">
     <p class="eyebrow">Ownership</p>
@@ -43,6 +77,8 @@
       <tbody>
         <tr><td>Project profile + lock</td><td>Project</td><td><span class="inline-code">.pi/piagent-profile*.json</span></td><td>Có</td></tr>
         <tr><td>Task, trace, telemetry, capture</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/</span></td><td>Không</td></tr>
+        <tr><td>Task journal + checkpoint</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/task-journal/</span></td><td>Không</td></tr>
+        <tr><td>Repository memory fact</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/repository-memory/</span></td><td>Không</td></tr>
         <tr><td>Session history</td><td>Pi</td><td>Pi session store</td><td>Không</td></tr>
         <tr><td>Shared instructions</td><td>Project</td><td><span class="inline-code">AGENTS.md</span>, project docs</td><td>Có</td></tr>
       </tbody>

package/docs-site/content/benchmark.html

@@ -1,7 +1,7 @@
 <section class="benchmark-summary" aria-labelledby="benchmark-verdict">
   <div class="benchmark-verdict">
     <div>
-      <p class="section-kicker">Production V1 · release evidence</p>
+      <p class="section-kicker">Production V1 · historical public regression evidence</p>
       <h2 id="benchmark-verdict">Piagent đạt 9,92/10</h2>
       <p>Hoàn thành toàn bộ 54 task, giữ Quality, Safety và Reliability ở 10/10, đồng thời dùng ít hơn 51,11% fresh token so với surface <code>codex-cli</code> trong phép đo ghép cặp.</p>
     </div>
@@ -182,6 +182,26 @@
   </div>
 </section>
 
+<section class="section">
+  <p class="section-kicker">Benchmark bands</p>
+  <h2 id="benchmark-bands">Chọn đúng suite theo quyết định cần đưa ra</h2>
+  <div class="table-wrap benchmark-table-wrap">
+    <table>
+      <thead><tr><th>Band</th><th>Khi nào dùng</th><th>Evidence tối thiểu</th><th>Claim được phép</th></tr></thead>
+      <tbody>
+        <tr><td>Core</td><td>Smoke nhanh trước khi làm release</td><td>4 scenario, 1 repeat, hidden verifier</td><td>Chỉ dùng để bắt regression</td></tr>
+        <tr><td>Production</td><td>Public regression cho release candidate</td><td>18 scenario, 3 repeat, <span class="inline-code">piagent</span> vs baseline</td><td>Claim chỉ giới hạn trong observed public suite; tách outcome coverage, comparable CI và failure-aware effort</td></tr>
+        <tr><td>Capability</td><td>Tìm trần năng lực của harness</td><td>4 bài lớn multi-file/multi-component với reference và mutation checks</td><td>Hill-climbing, không phải release hay generalization gate</td></tr>
+        <tr><td>Long-horizon</td><td>Thay đổi recovery/context và adopt repo lớn</td><td>Dedicated paid suite đang planned; hiện chạy Production cộng deterministic crash/replay tests</td><td>Chưa công bố điểm long-horizon cho tới khi suite được phát hành</td></tr>
+      </tbody>
+    </table>
+  </div>
+  <div class="callout">
+    <strong>Benchmark phải ghi đúng claim tier, không phải marketing copy.</strong>
+    <p>Một version không được claim tiết kiệm token nếu chưa pass quality/reliability, paired candidate-regression, outcome coverage, comparable-efficiency, failure-aware effort, scope/safety và protocol gate. So với raw Pi là causal harness ablation; so với <code>codex-cli</code> chỉ là external product reference.</p>
+  </div>
+</section>
+
 <section class="section">
   <p class="section-kicker">Provenance</p>
   <h2 id="minh-bach">Đọc kết quả đúng phạm vi</h2>
@@ -199,7 +219,7 @@
 
   <div class="callout">
     <strong>Giới hạn của kết luận</strong>
-    <p>Đây là synthetic production benchmark do maintainer xây dựng, không phải benchmark độc lập chứng minh Piagent tốt hơn trên mọi repository hoặc model. Raw report giữ private vì chứa seed tái tạo synthetic secret. Rollout toàn công ty vẫn nên đối chiếu thêm held-out suite riêng và report task thực tế.</p>
+    <p>Đây là public synthetic regression do maintainer xây dựng, không phải benchmark độc lập chứng minh Piagent tốt hơn trên mọi repository hoặc model. Generated values không tạo task family độc lập. Generalization cần external family-disjoint holdout, reference/mutation/human-calibration manifest, candidate freeze và shadow/canary task thực tế.</p>
   </div>
 
   <div class="command-grid">

package/docs-site/content/commands.html

@@ -8,8 +8,11 @@
         <span class="tag">/model</span>
         <span class="tag">/commands</span>
         <span class="tag">/piagent-status</span>
+        <span class="tag">/piagent-inspector</span>
         <span class="tag">/piagent-orchestration</span>
         <span class="tag">/usage</span>
+        <span class="tag">/usage efficiency</span>
+        <span class="tag">/task-preflight</span>
         <span class="tag">/context preflight</span>
         <span class="tag">/name</span>
         <span class="tag">/fresh</span>
@@ -82,3 +85,40 @@
     </article>
   </div>
 </section>
+
+<section class="section">
+  <p class="section-kicker">Product visibility</p>
+  <h2 id="product-status-command">Xem runtime đang quyết định gì</h2>
+  <p class="section-lead">Các command này đọc state deterministic, không cần model giải thích và không bắt buộc trong workflow hằng ngày.</p>
+  <div class="command-grid">
+    <article class="command-card">
+      <header><span>Preflight trước task</span><button class="clip-btn" data-clip="/task-preflight --json Review current repository state">Lấy lệnh</button></header>
+      <pre><code>/task-preflight --json Review current repository state</code></pre>
+      <p>Xem intent, risk, scope, runtime, solver route, phase, tool/helper recommendation, backend, approval và blocker trước khi bắt đầu.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Live task status</span><button class="clip-btn" data-clip="/piagent-status">Lấy lệnh</button></header>
+      <pre><code>/piagent-status</code></pre>
+      <p>Đọc Task Contract, checkpoint, verifier, recovery, helper, resume state và terminal receipt. Receipt chỉ approved khi hard completion gate trên working tree hiện tại pass.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Activity, diff và context</span><button class="clip-btn" data-clip="/piagent-inspector">Lấy lệnh</button></header>
+      <pre><code>/piagent-inspector</code></pre>
+      <p>Một menu duy nhất cho <span class="inline-code">summary</span>, <span class="inline-code">files</span>, <span class="inline-code">commands</span>, <span class="inline-code">security</span>, <span class="inline-code">context</span> và <span class="inline-code">toggle</span>. Panel bốn dòng tự hiện sát phía trên footer native; xanh lá là ổn, vàng cần chú ý, đỏ là failure/safety hoặc context từ 80%. Command chỉ đọc state local, không gọi model.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Efficiency có nguồn</span><button class="clip-btn" data-clip="/usage efficiency">Lấy lệnh</button></header>
+      <pre><code>/usage efficiency</code></pre>
+      <p>Xem context/tool utilization, verify, repair, retry và helper usage. Giá trị runtime không đo được giữ là <span class="inline-code">null</span>, không ước đoán.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Helper policy</span><button class="clip-btn" data-clip="/piagent-orchestration">Lấy lệnh</button></header>
+      <pre><code>/piagent-orchestration</code></pre>
+      <p>Xem solo-first policy, read-only roles, reviewer/Oracle recommendation, owned budget và one-writer invariant.</p>
+    </article>
+  </div>
+  <div class="callout">
+    <strong>Diff rõ nguồn, token không ước lượng.</strong>
+    <p>Footer status dùng file task đã quan sát để cập nhật nhanh; view <span class="inline-code">files</span> tính exact snapshot delta và đánh dấu <span class="inline-code">mixed-working-tree</span> nếu task đụng file đã dirty từ trước. Pi chỉ cung cấp model usage theo turn/session, nên token theo từng built-in tool luôn hiện unavailable.</p>
+  </div>
+</section>

package/docs-site/content/en/architecture.html

@@ -13,6 +13,22 @@
   </div>
 </section>
 
+<section class="section">
+  <div class="section-head">
+    <p class="eyebrow">Adaptive runtime</p>
+    <h2 id="adaptive-runtime">Piagent now separates runtime decisions from Pi wiring</h2>
+    <p>The harness keeps the model choice intact, but every task gets a smaller, cited context plan, durable task journal, and explicit recovery decision.</p>
+  </div>
+  <div class="grid cols-3">
+    <article class="card"><h3>Context planner</h3><p><span class="inline-code">runtime/context/adaptive-planner.ts</span> chooses budget, retrieval lane, code inclusion, and a plan receipt from phase, context pressure, model capability, and thinking level.</p></article>
+    <article class="card"><h3>Task journal</h3><p><span class="inline-code">extensions/task-journal.js</span> records start, progress, mutation, verifier, and completion checkpoints. Resume trusts the task contract first; retention compaction preserves hash anchors.</p></article>
+    <article class="card"><h3>Repository memory</h3><p><span class="inline-code">extensions/repository-memory.js</span> stores only cited, expiring facts and rejects raw prompts, outputs, OAuth material, API keys, and secrets. Hints use only unused context-plan budget.</p></article>
+    <article class="card"><h3>Model capability</h3><p><span class="inline-code">runtime/model/capabilities.ts</span> reads model family, context window, and thinking level as budget signals without switching providers or requiring API keys.</p></article>
+    <article class="card"><h3>Verification intelligence</h3><p><span class="inline-code">extensions/verification-intelligence.js</span> classifies failures and chooses a targeted verifier from task scope; mixed frontend/backend scope uses the full source verifier.</p></article>
+    <article class="card"><h3>Execution backend</h3><p><span class="inline-code">extensions/execution-backend.js</span> keeps local Pi OAuth as owner and blocks mutation when a requested experimental adapter is unavailable.</p></article>
+  </div>
+</section>
+
 <section class="section">
   <div class="section-head">
     <p class="eyebrow">Ownership</p>
@@ -25,6 +41,8 @@
       <tbody>
         <tr><td>Project profile and lock</td><td>Project</td><td><span class="inline-code">.pi/piagent-profile*.json</span></td><td>Yes</td></tr>
         <tr><td>Task, trace, telemetry, capture</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/</span></td><td>No</td></tr>
+        <tr><td>Task journal and checkpoints</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/task-journal/</span></td><td>No</td></tr>
+        <tr><td>Repository memory facts</td><td>Runtime</td><td><span class="inline-code">.pi/piagent-state/repository-memory/</span></td><td>No</td></tr>
         <tr><td>Session history</td><td>Pi</td><td>Pi session store</td><td>No</td></tr>
         <tr><td>Shared instructions</td><td>Project</td><td><span class="inline-code">AGENTS.md</span>, project docs</td><td>Yes</td></tr>
       </tbody>

package/docs-site/content/en/benchmark.html

@@ -1,7 +1,7 @@
 <section class="benchmark-summary" aria-labelledby="benchmark-verdict">
   <div class="benchmark-verdict">
     <div>
-      <p class="section-kicker">Production V1 · release evidence</p>
+      <p class="section-kicker">Production V1 · historical public regression evidence</p>
       <h2 id="benchmark-verdict">Piagent scores 9.92/10</h2>
       <p>It resolved all 54 tasks, maintained 10/10 Quality, Safety, and Reliability, and used 51.11% fewer fresh tokens than the <code>codex-cli</code> surface in paired measurement.</p>
     </div>
@@ -134,6 +134,26 @@
   </div>
 </section>
 
+<section class="section">
+  <p class="section-kicker">Benchmark bands</p>
+  <h2 id="benchmark-bands">Use the right suite for the decision</h2>
+  <div class="table-wrap benchmark-table-wrap">
+    <table>
+      <thead><tr><th>Band</th><th>When to use it</th><th>Minimum evidence</th><th>Claim allowed</th></tr></thead>
+      <tbody>
+        <tr><td>Core</td><td>Fast local smoke before release work</td><td>4 scenarios, 1 repeat, hidden verifier</td><td>Regression only</td></tr>
+        <tr><td>Production</td><td>Public regression for a release candidate</td><td>18 scenarios, 3 repeats, <span class="inline-code">piagent</span> vs baseline</td><td>Claims stay bounded to the observed public suite; outcome coverage, comparable CI, and failure-aware effort are separate</td></tr>
+        <tr><td>Capability</td><td>Find the harness capability ceiling</td><td>4 large multi-file/multi-component tasks with reference and mutation checks</td><td>Hill-climbing, not a release or generalization gate</td></tr>
+        <tr><td>Long-horizon</td><td>Recovery/context changes and large-repository adoption</td><td>Planned dedicated paid suite; today run Production plus deterministic crash/replay tests</td><td>No public long-horizon score until that suite ships</td></tr>
+      </tbody>
+    </table>
+  </div>
+  <div class="callout">
+    <strong>A benchmark must state its claim tier; it is not marketing copy.</strong>
+    <p>A version should not claim token savings unless quality/reliability, paired candidate-regression, outcome coverage, comparable-efficiency, failure-aware effort, scope/safety, and protocol gates pass. Raw Pi is a causal harness ablation; <code>codex-cli</code> is only an external product reference.</p>
+  </div>
+</section>
+
 <section class="section">
   <p class="section-kicker">Provenance</p>
   <h2 id="transparency">Interpret results within their measured scope</h2>
@@ -150,7 +170,7 @@
   </div>
   <div class="callout">
     <strong>Limits of the conclusion</strong>
-    <p>This is a synthetic production benchmark built by the maintainer, not an independent benchmark proving that Piagent is better for every repository or model. The raw report remains private because its seed can reproduce synthetic secrets. Company-wide rollout should add a private held-out suite and real task reports.</p>
+    <p>This is a maintainer-built public synthetic regression, not an independent benchmark proving Piagent is better for every repository or model. Generated values do not create independent task families. Generalization requires an external family-disjoint holdout, a reference/mutation/human-calibration manifest, a frozen candidate, and real shadow/canary tasks.</p>
   </div>
   <div class="command-grid">
     <div class="command-card">

package/docs-site/content/en/commands.html

@@ -2,7 +2,7 @@
   <div class="grid cols-2">
     <article class="card">
       <h3 id="session-basics">Session basics</h3>
-      <div class="tag-list"><span class="tag">/login</span><span class="tag">/model</span><span class="tag">/commands</span><span class="tag">/piagent-status</span><span class="tag">/piagent-orchestration</span><span class="tag">/usage</span><span class="tag">/context preflight</span><span class="tag">/name</span><span class="tag">/fresh</span><span class="tag">/session</span></div>
+      <div class="tag-list"><span class="tag">/login</span><span class="tag">/model</span><span class="tag">/commands</span><span class="tag">/piagent-status</span><span class="tag">/piagent-inspector</span><span class="tag">/piagent-orchestration</span><span class="tag">/usage</span><span class="tag">/usage efficiency</span><span class="tag">/task-preflight</span><span class="tag">/context preflight</span><span class="tag">/name</span><span class="tag">/fresh</span><span class="tag">/session</span></div>
       <p>Set up a session, select a model, inspect profile, usage, or orchestration state, and decide whether a large task needs a fresh session.</p>
     </article>
     <article class="card">
@@ -37,3 +37,40 @@
     </article>
   </div>
 </section>
+
+<section class="section">
+  <p class="section-kicker">Product visibility</p>
+  <h2 id="product-status-commands">Inspect runtime decisions</h2>
+  <p class="section-lead">These commands read deterministic state without asking a model to interpret it. They are optional in the daily workflow.</p>
+  <div class="command-grid">
+    <article class="command-card">
+      <header><span>Preflight before a task</span><button class="clip-btn" data-clip="/task-preflight --json Review current repository state">Copy</button></header>
+      <pre><code>/task-preflight --json Review current repository state</code></pre>
+      <p>Inspect intent, risk, scope, runtime, solver route, phases, tool and helper recommendations, backend, approvals, and blockers before starting.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Live task status</span><button class="clip-btn" data-clip="/piagent-status">Copy</button></header>
+      <pre><code>/piagent-status</code></pre>
+      <p>Read the Task Contract, checkpoints, verifier evidence, recovery, helpers, resume state, and terminal receipt. Approval requires the current-tree hard completion gate to pass.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Activity, diff, and context</span><button class="clip-btn" data-clip="/piagent-inspector">Copy</button></header>
+      <pre><code>/piagent-inspector</code></pre>
+      <p>One menu for <span class="inline-code">summary</span>, <span class="inline-code">files</span>, <span class="inline-code">commands</span>, <span class="inline-code">security</span>, <span class="inline-code">context</span>, and <span class="inline-code">toggle</span>. A four-row panel appears automatically just above the native footer: green is healthy, yellow needs attention, and red marks failures, safety warnings, or context at 80% and above. The command reads local state without calling a model.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Sourced efficiency</span><button class="clip-btn" data-clip="/usage efficiency">Copy</button></header>
+      <pre><code>/usage efficiency</code></pre>
+      <p>Inspect context and tool utilization, verification, repair, retries, and helper usage. Runtime values that cannot be measured stay <span class="inline-code">null</span>.</p>
+    </article>
+    <article class="command-card">
+      <header><span>Helper policy</span><button class="clip-btn" data-clip="/piagent-orchestration">Copy</button></header>
+      <pre><code>/piagent-orchestration</code></pre>
+      <p>Inspect the solo-first policy, read-only roles, reviewer or Oracle recommendations, owned budgets, and the one-writer invariant.</p>
+    </article>
+  </div>
+  <div class="callout">
+    <strong>Sourced diffs, no token estimates.</strong>
+    <p>The footer panel uses observed task files for fast updates; the <span class="inline-code">files</span> view computes the exact snapshot delta and marks <span class="inline-code">mixed-working-tree</span> when a task touches a file that was already dirty. Pi exposes model usage by turn and session, so per-built-in-tool tokens remain unavailable.</p>
+  </div>
+</section>

package/docs-site/content/en/index.html

@@ -12,6 +12,7 @@
   </div>
   <div class="hero-actions">
     <a class="button primary" href="/en/quickstart">Get started</a>
+    <a class="button" href="/en/whats-new">What’s new</a>
     <a class="button" href="/en/commands">View commands</a>
     <a class="button" href="/en/mcp">MCP</a>
   </div>
@@ -28,6 +29,11 @@ piagent-setup</code></pre>
 <section class="section">
   <h2 id="where-to-start">Where to start</h2>
   <div class="card-grid">
+    <a class="doc-card" href="/en/whats-new">
+      <span class="card-kicker">Product update</span>
+      <h3>What’s new</h3>
+      <p>Runtime truth, task trajectories, recovery, helper policy, and fail-closed completion receipts — including what runs automatically and what is opt-in.</p>
+    </a>
     <a class="doc-card" href="/en/quickstart">
       <span class="card-kicker">Get started</span>
       <h3>Quickstart</h3>

package/docs-site/content/en/runtime.html

@@ -1,7 +1,7 @@
 <section class="section">
   <div class="command-grid">
     <article class="command-card"><header><span>Global setup</span><button class="clip-btn" data-clip="piagent-setup --mcp-preset core --subagents-preset safe">Copy</button></header><pre><code>piagent-setup --mcp-preset core --subagents-preset safe</code></pre></article>
-    <article class="command-card"><header><span>Full update</span><button class="clip-btn" data-clip="npm install -g --ignore-scripts @earendil-works/pi-coding-agent@0.82.0&#10;npm install -g --ignore-scripts @piagent/platform@X.Y.Z&#10;piagent-install --stable --dry-run&#10;piagent-install --stable">Copy</button></header><pre><code>npm install -g --ignore-scripts @earendil-works/pi-coding-agent@0.82.0
+    <article class="command-card"><header><span>Full update</span><button class="clip-btn" data-clip="npm install -g --ignore-scripts @earendil-works/pi-coding-agent@0.84.1&#10;npm install -g --ignore-scripts @piagent/platform@X.Y.Z&#10;piagent-install --stable --dry-run&#10;piagent-install --stable">Copy</button></header><pre><code>npm install -g --ignore-scripts @earendil-works/pi-coding-agent@0.84.1
 npm install -g --ignore-scripts @piagent/platform@X.Y.Z
 piagent-install --stable --dry-run
 piagent-install --stable</code></pre></article>