@netlify/axis 0.2.0 → 0.3.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@netlify/axis",
-  "version": "0.2.0",
+  "version": "0.3.0",
   "description": "Agent eXperience Index Score — synthetic testing for AI agent interaction",
   "type": "module",
   "main": "./dist/index.js",
@@ -26,7 +26,7 @@
   },
   "lint-staged": {
     "*.{ts,tsx}": [
-      "eslint --fix",
+      "eslint --fix --no-warn-ignored",
       "prettier --write"
     ],
     "*.{json,md,yml,yaml}": [
@@ -37,7 +37,9 @@
     "node": ">=18.0.0"
   },
   "files": [
-    "dist"
+    "dist",
+    "!dist/docs-site",
+    "!dist/report-ui/mock-data.json"
   ],
   "devDependencies": {
     "@eslint/js": "^9.39.4",

package/dist/docs-site/_astro/cli.DDWZtG0-.css

@@ -1 +0,0 @@
-:root{--bg: #fafbf9;--bg-card: #ffffff;--text: #1a1a2e;--text-secondary: #4a5568;--text-muted: #637083;--border: #e5e7eb;--border-bright: #d1d5db;--light-gray: #f3f4f6;--accent: #016867;--accent-hover: #015554;--accent-bright: #2dd4bf;--accent-light: rgba(1, 104, 103, .07);--color-env: #059669;--color-svc: #2563eb;--color-agent: #6b7280;--warning: #d97706;--warning-light: rgba(217, 119, 6, .06);--danger: #dc2626;--danger-light: rgba(220, 38, 38, .06);--success: #059669;--success-light: rgba(5, 150, 105, .06);--sidebar-bg: #0f2b2b;--sidebar-text: #8faca8;--sidebar-text-muted: #7a9b97;--sidebar-hover: rgba(255, 255, 255, .06);--sidebar-active-bg: rgba(45, 212, 191, .12);--sidebar-active-text: #2dd4bf;--radius: 8px;--radius-lg: 12px;--shadow: 0 1px 3px rgba(0, 0, 0, .06);--shadow-lg: 0 8px 24px rgba(0, 0, 0, .08);--glow: 0 0 40px rgba(1, 104, 103, .06);--font: "Inter", -apple-system, BlinkMacSystemFont, "Segoe UI", system-ui, sans-serif;--font-mono: "SF Mono", SFMono-Regular, Menlo, Consolas, monospace;--transition: .15s ease;--sidebar-width: 240px}*,*:before,*:after{box-sizing:border-box;margin:0;padding:0}html{font-size:15px;-webkit-font-smoothing:antialiased;-moz-osx-font-smoothing:grayscale}body{font-family:var(--font);color:var(--text);background:var(--bg);background-image:radial-gradient(ellipse at top,rgba(1,104,103,.04) 0%,transparent 60%);background-attachment:fixed;line-height:1.6;min-height:100vh}:focus-visible{outline:2px solid var(--accent);outline-offset:2px;border-radius:2px}.site-header{position:sticky;top:0;z-index:50;background:#fafbf9d9;backdrop-filter:blur(16px);-webkit-backdrop-filter:blur(16px);border-bottom:1px solid var(--border);padding:0 24px;height:56px;display:flex;align-items:center;justify-content:space-between}.site-logo{display:flex;align-items:center;gap:10px;text-decoration:none;color:var(--text)}.site-logo-mark{font-size:2.0625rem;font-weight:800;letter-spacing:.12em;color:var(--accent)}.logo-ax{letter-spacing:-.06em}.logo-i{display:inline-block;font-style:italic;transform:skew(-20deg);margin-left:2px;margin-right:-2px}.site-header-links{display:flex;gap:16px;align-items:center}.site-header-links a{font-size:1rem;color:var(--text-secondary);text-decoration:none;font-weight:500;transition:color var(--transition)}.site-header-links a:hover{color:var(--accent)}.mobile-menu-btn{display:none;background:none;border:none;cursor:pointer;color:var(--text);font-size:1.25rem;padding:4px}.site-shell{display:flex;min-height:calc(100vh - 56px)}.sidebar{width:var(--sidebar-width);flex-shrink:0;background:radial-gradient(ellipse at 20% 15%,rgba(45,212,191,.18) 0%,transparent 50%),radial-gradient(ellipse at 80% 85%,rgba(30,80,160,.2) 0%,transparent 50%),radial-gradient(ellipse at 50% 50%,rgba(14,74,74,.4) 0%,transparent 70%),linear-gradient(180deg,#0d3d3d,#0a2424);padding:24px 0;position:sticky;top:56px;height:calc(100vh - 56px);overflow-y:auto}.sidebar-nav{display:flex;flex-direction:column;gap:2px;padding:0 12px}.sidebar-link{display:block;padding:8px 16px;font-size:.875rem;font-weight:500;color:var(--sidebar-text);text-decoration:none;border-radius:var(--radius);transition:background var(--transition),color var(--transition)}.sidebar-link:hover{background:var(--sidebar-hover);color:#c8dbd9}.sidebar-link.active{background:var(--sidebar-active-bg);color:var(--sidebar-active-text);font-weight:600}.sidebar-section-label{padding:20px 16px 6px;font-size:.6875rem;font-weight:600;text-transform:uppercase;letter-spacing:.06em;color:var(--sidebar-text-muted)}.main-content{flex:1;min-width:0;padding:40px 48px 80px;max-width:860px}h1{font-size:2.25rem;font-weight:800;letter-spacing:-.03em;line-height:1.15;margin-bottom:12px;background:linear-gradient(135deg,var(--text) 40%,var(--accent) 100%);-webkit-background-clip:text;-webkit-text-fill-color:transparent;background-clip:text}h2{font-size:1.5rem;font-weight:700;letter-spacing:-.02em;color:var(--text);margin-top:48px;margin-bottom:16px;padding-bottom:12px;border-bottom:none;background-image:linear-gradient(90deg,rgba(1,104,103,.35),transparent 80%);background-size:100% 1px;background-position:bottom left;background-repeat:no-repeat}h3{font-size:1.125rem;font-weight:600;color:var(--text);margin-top:32px;margin-bottom:12px}h4{font-size:.9375rem;font-weight:600;color:var(--text-secondary);margin-top:24px;margin-bottom:8px}p{margin-bottom:16px;color:var(--text-secondary)}.lead{font-size:1.125rem;color:var(--text-secondary);line-height:1.7;margin-bottom:32px}a{color:var(--accent);text-decoration:underline;text-decoration-thickness:1px;text-underline-offset:2px;transition:color var(--transition)}a:hover{color:var(--accent-hover)}.sidebar-link,.site-logo,.site-header-links a{text-decoration:none}strong{font-weight:600;color:var(--text)}ul,ol{margin-bottom:16px;padding-left:24px}li{margin-bottom:6px;color:var(--text-secondary)}li>ul,li>ol{margin-top:6px;margin-bottom:0}code{font-family:var(--font-mono);font-size:.8125em;background:var(--accent-light);padding:2px 6px;border-radius:4px;color:var(--accent)}pre{background:#0f2b2b;color:#d4e8e6;padding:20px 24px;border-radius:var(--radius-lg);overflow-x:auto;margin-bottom:24px;font-size:.8125rem;line-height:1.7;border:1px solid rgba(255,255,255,.08);border-left:3px solid rgba(45,212,191,.5);box-shadow:var(--glow)}pre code{background:none;padding:0;border-radius:0;color:inherit;font-size:inherit}.card-grid{display:grid;grid-template-columns:repeat(2,1fr);gap:16px;margin-bottom:32px}.card{background:var(--bg-card);border:1px solid var(--border);border-radius:var(--radius-lg);padding:20px;box-shadow:var(--shadow);transition:border-color var(--transition),box-shadow var(--transition)}.card:hover{border-color:var(--border-bright);box-shadow:0 4px 24px #01686714,var(--glow)}.card-title{font-size:.9375rem;font-weight:600;margin-bottom:6px;color:var(--text)}.card-desc{font-size:.8125rem;color:var(--text-secondary);line-height:1.5}.card-accent{border-top:3px solid var(--accent)}.card-env{border-top:3px solid var(--color-env)}.card-svc{border-top:3px solid var(--color-svc)}.card-agent{border-top:3px solid var(--color-agent)}.card-weight{display:inline-block;font-size:.6875rem;font-weight:700;font-family:var(--font-mono);background:var(--accent-light);color:var(--accent);padding:2px 8px;border-radius:10px;margin-bottom:8px}table{width:100%;border-collapse:collapse;margin-bottom:24px;font-size:.875rem}thead th{text-align:left;padding:10px 12px;font-size:.75rem;text-transform:uppercase;letter-spacing:.06em;color:var(--text-muted);font-weight:600;border-bottom:2px solid var(--border)}tbody td{padding:10px 12px;border-bottom:1px solid var(--light-gray);vertical-align:top;color:var(--text-secondary)}tbody tr:hover{background:var(--light-gray)}td code{font-size:.75rem}.callout{padding:16px 20px;border-radius:var(--radius);margin-bottom:24px;font-size:.875rem;line-height:1.6}.callout p:last-child{margin-bottom:0}.callout-title{font-weight:600;font-size:.75rem;text-transform:uppercase;letter-spacing:.04em;margin-bottom:4px}.callout-info{background:var(--accent-light);border-left:3px solid var(--accent)}.callout-info .callout-title{color:var(--accent)}.callout-warn{background:var(--warning-light);border-left:3px solid var(--warning)}.callout-warn .callout-title{color:var(--warning)}.callout-success{background:var(--success-light);border-left:3px solid var(--success)}.callout-success .callout-title{color:var(--success)}.pipeline{display:flex;flex-wrap:wrap;gap:8px;align-items:center;margin-bottom:32px;padding:24px;background:var(--bg-card);border:1px solid var(--border);border-radius:var(--radius-lg);box-shadow:var(--shadow)}.pipeline-step{display:flex;align-items:center;gap:4px;padding:6px 14px;background:var(--light-gray);border:1px solid var(--border);border-radius:20px;font-size:.8125rem;font-weight:500;color:var(--text-secondary);white-space:nowrap}.pipeline-step.llm{background:var(--accent-light);border-color:var(--accent);color:var(--accent);font-weight:600}.pipeline-arrow{color:var(--text-muted);font-size:.875rem;flex-shrink:0}.hero{margin-bottom:48px}.hero-badge{display:inline-block;font-size:.6875rem;font-weight:600;text-transform:uppercase;letter-spacing:.06em;color:var(--accent);background:var(--accent-light);border:1px solid rgba(1,104,103,.2);padding:4px 12px;border-radius:12px;margin-bottom:16px}.cli-block{background:var(--bg-card);border:1px solid var(--border);border-radius:var(--radius-lg);padding:20px 24px;margin-bottom:16px;box-shadow:var(--shadow);transition:border-color var(--transition),box-shadow var(--transition)}.cli-block:hover{border-color:var(--border-bright);box-shadow:var(--shadow),var(--glow)}.cli-command{font-family:var(--font-mono);font-size:.9375rem;font-weight:600;color:var(--accent);margin-bottom:8px}.cli-desc{font-size:.875rem;color:var(--text-secondary);margin-bottom:12px}.cli-flags{display:flex;flex-direction:column;gap:6px}.cli-flag{display:flex;gap:12px;font-size:.8125rem}.cli-flag-name{font-family:var(--font-mono);font-weight:500;color:var(--text);flex-shrink:0;min-width:200px}.cli-flag-desc{color:var(--text-secondary)}.site-footer{margin-top:64px;padding:24px 0;border-top:1px solid var(--border);font-size:.8125rem;color:var(--text-muted)}.dimension-wheel{display:grid;grid-template-columns:1fr 200px 1fr;grid-template-rows:1fr 1fr;gap:20px 28px;align-items:center;margin-bottom:32px;padding:24px 0}.dimension-chart{grid-column:2;grid-row:1 / 3;width:200px;height:200px}.dimension-chart circle{transition:stroke-width .2s ease,opacity .2s ease;cursor:pointer}.dimension-chart text{transition:opacity .2s ease}.dimension-label{display:block;padding:10px 14px;border-radius:var(--radius);text-decoration:none;transition:background var(--transition),box-shadow var(--transition)}.dimension-label:hover{background:var(--light-gray);box-shadow:var(--shadow)}.dimension-pct{display:block;font-size:.75rem;font-weight:700;font-family:var(--font-mono);margin-bottom:2px}.dimension-name{display:block;font-size:.9375rem;font-weight:600;color:var(--text);margin-bottom:2px}.dimension-desc{display:block;font-size:.8125rem;color:var(--text-secondary);line-height:1.4}.dimension-label-goal{grid-column:3;grid-row:1;justify-self:start}.dimension-label-env{grid-column:3;grid-row:2;justify-self:start}.dimension-label-svc{grid-column:1;grid-row:2;justify-self:end;text-align:right}.dimension-label-agent{grid-column:1;grid-row:1;justify-self:end;text-align:right}@media(max-width:768px){.mobile-menu-btn{display:block}.sidebar{position:fixed;top:56px;left:0;bottom:0;z-index:40;transform:translate(-100%);transition:transform .2s ease;box-shadow:var(--shadow-lg)}.sidebar.open{transform:translate(0)}.sidebar-backdrop{display:none;position:fixed;inset:56px 0 0;z-index:39;background:#0f2b2b80}.sidebar-backdrop.open{display:block}.main-content{padding:24px 16px 64px}.card-grid{grid-template-columns:1fr}.dimension-wheel{grid-template-columns:1fr;grid-template-rows:auto;justify-items:center}.dimension-chart{grid-column:1;grid-row:1}.dimension-label-goal{grid-column:1;grid-row:2;justify-self:stretch;text-align:left}.dimension-label-env{grid-column:1;grid-row:3;justify-self:stretch;text-align:left}.dimension-label-svc{grid-column:1;grid-row:4;justify-self:stretch;text-align:left}.dimension-label-agent{grid-column:1;grid-row:5;justify-self:stretch;text-align:left}.pipeline{flex-direction:column;align-items:stretch}.pipeline-arrow{text-align:center;transform:rotate(90deg)}.cli-flag{flex-direction:column;gap:2px}.cli-flag-name{min-width:0}}

package/dist/docs-site/cli/index.html

@@ -1,18 +0,0 @@
-<!DOCTYPE html><html lang="en"> <head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>CLI - AXIS Docs</title><meta name="description" content="Documentation for AXIS, the Agent eXperience Index Score synthetic testing framework for AI agents."><link rel="stylesheet" href="/_astro/cli.DDWZtG0-.css"></head> <body> <header class="site-header"> <a href="/" class="site-logo" aria-label="AXIS home"> <span class="site-logo-mark"><span class="logo-ax">AX</span><span class="logo-i">I</span>S</span> </a> <div class="site-header-links"> <a href="https://github.com/netlify/axis">GitHub</a> <a href="https://www.npmjs.com/package/@netlify/axis">npm</a> </div> <button class="mobile-menu-btn" id="menu-btn" aria-label="Toggle navigation">&#9776;</button> </header> <div class="site-shell"> <div class="sidebar-backdrop" id="sidebar-backdrop"></div> <aside class="sidebar" id="sidebar"> <nav class="sidebar-nav"> <div class="sidebar-section-label">Getting Started</div> <a href="/" class="sidebar-link">What is AXIS</a> <a href="/quickstart" class="sidebar-link">Quick Start</a> <div class="sidebar-section-label">How It Works</div> <a href="/scoring" class="sidebar-link">Scoring Framework</a> <a href="/running" class="sidebar-link">Running Tests</a> <div class="sidebar-section-label">Reference</div> <a href="/cli" class="sidebar-link active">CLI</a> <a href="/configuration" class="sidebar-link">Configuration</a> </nav> </aside> <main class="main-content">  <h1>CLI Reference</h1> <p class="lead">
-Complete reference for the <code>axis</code> command-line interface. All commands can be run
-    directly or via <code>npx @netlify/axis</code>.
-</p> <h2><code>axis run</code></h2> <div class="cli-block"> <div class="cli-command">axis run [options]</div> <div class="cli-desc">Execute scenarios against configured agents.</div> <div class="cli-flags"> <div class="cli-flag"> <span class="cli-flag-name"><code>-c, --config &lt;path&gt;</code></span> <span class="cli-flag-desc">Config file (default: <code>axis.config.json</code>).</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-s, --scenario &lt;key&gt;</code></span> <span class="cli-flag-desc">Run a specific scenario by key.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-a, --agent &lt;name&gt;</code></span> <span class="cli-flag-desc">Run with a specific agent only.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--json</code></span> <span class="cli-flag-desc">JSON output to stdout (no live terminal display).</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-v, --verbose</code></span> <span class="cli-flag-desc">Detailed per-step logging.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-o, --output-dir &lt;dir&gt;</code></span> <span class="cli-flag-desc">Also write report files to this directory.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--concurrency &lt;n&gt;</code></span> <span class="cli-flag-desc">Max parallel jobs.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--debug</code></span> <span class="cli-flag-desc">Capture raw agent stdout for debugging.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--no-score</code></span> <span class="cli-flag-desc">Skip scoring (raw results only).</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--refresh-skills</code></span> <span class="cli-flag-desc">Force re-clone remote skills.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--compare-baseline [name]</code></span> <span class="cli-flag-desc">Diff against a baseline after scoring.</span> </div> </div> </div> <h2><code>axis reports</code></h2> <div class="cli-block"> <div class="cli-command">axis reports [reportId] [scenarioKey] [options]</div> <div class="cli-desc">View past AXIS reports.</div> <div class="cli-flags"> <div class="cli-flag"> <span class="cli-flag-name"><code>[reportId]</code></span> <span class="cli-flag-desc">Report ID or <code>latest</code>. Omit to list all reports.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>[scenarioKey]</code></span> <span class="cli-flag-desc">Drill into a specific scenario detail.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-a, --agent &lt;name...&gt;</code></span> <span class="cli-flag-desc">Filter by agent(s), repeatable.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--json</code></span> <span class="cli-flag-desc">Output as JSON.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--html</code></span> <span class="cli-flag-desc">Open report in browser.</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>-n, --limit &lt;count&gt;</code></span> <span class="cli-flag-desc">Max reports to list (default: 10).</span> </div> </div> </div> <h2><code>axis baseline</code></h2> <div class="cli-block"> <div class="cli-command">axis baseline set [name]</div> <div class="cli-desc">Create or update a baseline snapshot from a report.</div> <div class="cli-flags"> <div class="cli-flag"> <span class="cli-flag-name"><code>--from &lt;reportId&gt;</code></span> <span class="cli-flag-desc">Use a specific report (default: latest).</span> </div> </div> </div> <div class="cli-block"> <div class="cli-command">axis baseline list</div> <div class="cli-desc">List all saved baselines.</div> </div> <div class="cli-block"> <div class="cli-command">axis baseline show [name]</div> <div class="cli-desc">Display baseline contents.</div> <div class="cli-flags"> <div class="cli-flag"> <span class="cli-flag-name"><code>--json</code></span> <span class="cli-flag-desc">Output as JSON.</span> </div> </div> </div> <div class="cli-block"> <div class="cli-command">axis baseline diff [name]</div> <div class="cli-desc">Compare a report against a baseline. Exits with code 1 if regressions are detected.</div> <div class="cli-flags"> <div class="cli-flag"> <span class="cli-flag-name"><code>--report &lt;reportId&gt;</code></span> <span class="cli-flag-desc">Specific report to compare (default: latest).</span> </div> <div class="cli-flag"> <span class="cli-flag-name"><code>--json</code></span> <span class="cli-flag-desc">Output as JSON.</span> </div> </div> </div> <div class="cli-block"> <div class="cli-command">axis baseline delete [name]</div> <div class="cli-desc">Delete a saved baseline.</div> </div>  <footer class="site-footer">
-AXIS is maintained by Netlify.
-</footer> </main> </div> <script>
-      const btn = document.getElementById("menu-btn");
-      const sidebar = document.getElementById("sidebar");
-      const backdrop = document.getElementById("sidebar-backdrop");
-
-      function toggle() {
-        sidebar.classList.toggle("open");
-        backdrop.classList.toggle("open");
-      }
-
-      btn.addEventListener("click", toggle);
-      backdrop.addEventListener("click", toggle);
-    </script> </body> </html>

package/dist/docs-site/configuration/index.html

@@ -1,121 +0,0 @@
-<!DOCTYPE html><html lang="en"> <head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>Configuration - AXIS Docs</title><meta name="description" content="Documentation for AXIS, the Agent eXperience Index Score synthetic testing framework for AI agents."><link rel="stylesheet" href="/_astro/cli.DDWZtG0-.css"></head> <body> <header class="site-header"> <a href="/" class="site-logo" aria-label="AXIS home"> <span class="site-logo-mark"><span class="logo-ax">AX</span><span class="logo-i">I</span>S</span> </a> <div class="site-header-links"> <a href="https://github.com/netlify/axis">GitHub</a> <a href="https://www.npmjs.com/package/@netlify/axis">npm</a> </div> <button class="mobile-menu-btn" id="menu-btn" aria-label="Toggle navigation">&#9776;</button> </header> <div class="site-shell"> <div class="sidebar-backdrop" id="sidebar-backdrop"></div> <aside class="sidebar" id="sidebar"> <nav class="sidebar-nav"> <div class="sidebar-section-label">Getting Started</div> <a href="/" class="sidebar-link">What is AXIS</a> <a href="/quickstart" class="sidebar-link">Quick Start</a> <div class="sidebar-section-label">How It Works</div> <a href="/scoring" class="sidebar-link">Scoring Framework</a> <a href="/running" class="sidebar-link">Running Tests</a> <div class="sidebar-section-label">Reference</div> <a href="/cli" class="sidebar-link">CLI</a> <a href="/configuration" class="sidebar-link active">Configuration</a> </nav> </aside> <main class="main-content">  <h1>Configuration Reference</h1> <p class="lead">
-Complete reference for <code>axis.config.json</code> and scenario files.
-</p> <h2>axis.config.json</h2> <p>
-AXIS is configured via an <code>axis.config.json</code> file in your project root.
-</p> <pre><code>{
-  &quot;scenarios&quot;: &quot;./scenarios&quot;,
-  &quot;agents&quot;: [
-    &quot;claude-code&quot;,
-    {
-      &quot;adapter&quot;: &quot;gemini&quot;,
-      &quot;model&quot;: &quot;gemini-2.5-pro&quot;,
-      &quot;scenarios&quot;: [&quot;cms/*&quot;],
-      &quot;flags&quot;: { &quot;yolo&quot;: true }
-    }
-  ],
-  &quot;defaults&quot;: {
-    &quot;concurrency&quot;: 4,
-    &quot;scoring_weights&quot;: {
-      &quot;goal_achievement&quot;: 0.4,
-      &quot;environment&quot;: 0.2,
-      &quot;service&quot;: 0.2,
-      &quot;agent&quot;: 0.2
-    }
-  },
-  &quot;env&quot;: [&quot;ANTHROPIC_API_KEY&quot;, &quot;GEMINI_API_KEY&quot;],
-  &quot;mcp_servers&quot;: {
-    &quot;filesystem&quot;: {
-      &quot;type&quot;: &quot;stdio&quot;,
-      &quot;command&quot;: &quot;npx&quot;,
-      &quot;args&quot;: [&quot;-y&quot;, &quot;@modelcontextprotocol/server-filesystem&quot;, &quot;/tmp&quot;]
-    }
-  },
-  &quot;skills&quot;: [&quot;./skills/deploy&quot;],
-  &quot;adapters&quot;: {
-    &quot;my-agent&quot;: &quot;./adapters/my-agent.ts&quot;
-  }
-}</code></pre> <h3>Top-Level Fields</h3> <table> <thead> <tr> <th>Field</th> <th>Type</th> <th>Description</th> </tr> </thead> <tbody> <tr> <td><code>scenarios</code></td> <td><code>string</code></td> <td>Path to the scenarios directory (required).</td> </tr> <tr> <td><code>agents</code></td> <td><code>(string | AgentConfig)[]</code></td> <td>Adapter names or full agent configurations (required).</td> </tr> <tr> <td><code>defaults</code></td> <td><code>object</code></td> <td>Default concurrency and scoring weights.</td> </tr> <tr> <td><code>env</code></td> <td><code>string[]</code></td> <td>Environment variables to pass through to agent processes.</td> </tr> <tr> <td><code>mcp_servers</code></td> <td><code>object</code></td> <td>MCP servers available to all agents.</td> </tr> <tr> <td><code>skills</code></td> <td><code>string[]</code></td> <td>Skills available to all agents (local paths, GitHub shorthand, or URLs).</td> </tr> <tr> <td><code>adapters</code></td> <td><code>object</code></td> <td>Custom adapter module paths, keyed by adapter name.</td> </tr> </tbody> </table> <h3>Agent Configuration</h3> <p>
-Agents can be specified as a simple string (adapter name) or a full configuration object.
-</p> <table> <thead> <tr> <th>Field</th> <th>Type</th> <th>Description</th> </tr> </thead> <tbody> <tr> <td><code>adapter</code></td> <td><code>string</code></td> <td>Adapter type (required): <code>claude-code</code>, <code>codex</code>, <code>gemini</code>, etc.</td> </tr> <tr> <td><code>model</code></td> <td><code>string</code></td> <td>Model override passed to the agent CLI.</td> </tr> <tr> <td><code>scenarios</code></td> <td><code>string[]</code></td> <td>Subset of scenarios to run (supports glob patterns like <code>cms/*</code>).</td> </tr> <tr> <td><code>skills</code></td> <td><code>string[]</code></td> <td>Agent-specific skills (merged with top-level skills).</td> </tr> <tr> <td><code>flags</code></td> <td><code>object</code></td> <td>CLI flags passed to the agent (for example, <code>{&quot;full-auto&quot;: true}</code>).</td> </tr> <tr> <td><code>command</code></td> <td><code>string</code></td> <td>Custom CLI command (for custom adapters).</td> </tr> </tbody> </table><code> <h3>Scoring Weights</h3> <p>
-Override the default dimension weights under <code>defaults.scoring_weights</code>. Values
-    must sum to 1.0.
-</p> <pre><code>{
-  &quot;defaults&quot;: {
-    &quot;scoring_weights&quot;: {
-      &quot;goal_achievement&quot;: 0.4,
-      &quot;environment&quot;: 0.2,
-      &quot;service&quot;: 0.2,
-      &quot;agent&quot;: 0.2
-    }
-  }
-}</code></pre> <h3>MCP Servers</h3> <p>
-Configure Model Context Protocol servers that are automatically wired into each agent
-    environment. AXIS supports both stdio (local process) and HTTP (remote endpoint) servers.
-</p> <pre><code>{
-  &quot;mcp_servers&quot;: {
-    &quot;filesystem&quot;: {
-      &quot;type&quot;: &quot;stdio&quot;,
-      &quot;command&quot;: &quot;npx&quot;,
-      &quot;args&quot;: [&quot;-y&quot;, &quot;@modelcontextprotocol/server-filesystem&quot;, &quot;/tmp&quot;],
-      &quot;env&quot;: { &quot;LOG_LEVEL&quot;: &quot;info&quot; }
-    },
-    &quot;remote-api&quot;: {
-      &quot;type&quot;: &quot;http&quot;,
-      &quot;url&quot;: &quot;https://mcp.example.com/tools&quot;,
-      &quot;headers&quot;: { &quot;Authorization&quot;: &quot;Bearer ${TOKEN}&quot; }
-    }
-  }
-}</code></pre> <p>
-Each adapter writes MCP configuration in its native format before spawning the agent.
-</p> <ul> <li><strong>Claude Code:</strong> <code>.mcp.json</code> in workspace root.</li> <li><strong>Codex:</strong> <code>config.toml</code> in <code>CODEX_HOME</code>.</li> <li><strong>Gemini:</strong> <code>settings.json</code> in <code>GEMINI_CLI_HOME</code>.</li> </ul> <h3>Skills</h3> <p>
-Skills extend agent capabilities with reusable instruction sets. Specify them at the top level
-    (shared), per agent, or per scenario.
-</p> <pre><code>{
-  &quot;skills&quot;: [
-    &quot;./skills/deploy&quot;,
-    &quot;netlify/axis-skill-deploy&quot;,
-    &quot;https://github.com/owner/repo&quot;
-  ]
-}</code></pre> <ul> <li><strong>Local paths:</strong> Relative to the config file.</li> <li><strong>GitHub shorthand:</strong> <code>owner/repo</code> format.</li> <li><strong>Full URLs:</strong> GitHub repository URLs.</li> </ul> <p>
-Remote skills are cached in <code>.axis/skills-cache/</code>. Use <code>--refresh-skills</code>
-to force re-clone.
-</p> <h3>Environment Variables</h3> <p>
-The <code>env</code> field lists additional environment variables to pass through to agent
-    processes. The following are always passed through by default:
-</p> <ul> <li><code>ANTHROPIC_API_KEY</code>, <code>CODEX_API_KEY</code>, <code>GEMINI_API_KEY</code></li> <li>System essentials: <code>PATH</code>, <code>USER</code>, <code>SHELL</code>, <code>LANG</code>, <code>TERM</code>, <code>TMPDIR</code></li> </ul> <h2>Scenarios</h2> <p>
-Scenarios are JSON files in the configured scenarios directory. The filename (without <code>.json</code>)
-    becomes the scenario key. Nested directories create namespaced keys.
-</p> <ul> <li><code>scenarios/hello-world.json</code> maps to key <code>hello-world</code>.</li> <li><code>scenarios/cms/create-post.json</code> maps to key <code>cms/create-post</code>.</li> </ul> <h3>Scenario Schema</h3> <pre><code>{
-  &quot;name&quot;: &quot;Debug and fix a broken script&quot;,
-  &quot;prompt&quot;: &quot;There is a JavaScript file at /tmp/app/add.js that has a bug. Find it, fix it, and verify.&quot;,
-  &quot;rubric&quot;: [
-    { &quot;check&quot;: &quot;Agent identified the bug&quot;, &quot;weight&quot;: 0.3 },
-    { &quot;check&quot;: &quot;Agent fixed the bug&quot;, &quot;weight&quot;: 0.4 },
-    { &quot;check&quot;: &quot;Agent verified the fix&quot;, &quot;weight&quot;: 0.3 }
-  ],
-  &quot;setup&quot;: [
-    { &quot;action&quot;: &quot;run_script&quot;, &quot;command&quot;: &quot;mkdir -p /tmp/app&quot; },
-    { &quot;action&quot;: &quot;run_script&quot;, &quot;command&quot;: &quot;echo &#39;function add(a,b) { return a-b; }&#39; &gt; /tmp/app/add.js&quot; }
-  ],
-  &quot;teardown&quot;: [
-    { &quot;action&quot;: &quot;run_script&quot;, &quot;command&quot;: &quot;rm -rf /tmp/app&quot; }
-  ],
-  &quot;agents&quot;: [&quot;claude-code&quot;]
-}</code></pre> <table> <thead> <tr> <th>Field</th> <th>Type</th> <th>Description</th> </tr> </thead> <tbody> <tr> <td><code>name</code></td> <td><code>string</code></td> <td>Human-readable scenario title (required).</td> </tr> <tr> <td><code>prompt</code></td> <td><code>string</code></td> <td>Task description sent to the agent (required).</td> </tr> <tr> <td><code>rubric</code></td> <td><code>string | object[]</code></td> <td>Success criteria: a string or array of checks with optional weights (required).</td> </tr> <tr> <td><code>setup</code></td> <td><code>object[]</code></td> <td>Lifecycle actions run before the agent.</td> </tr> <tr> <td><code>teardown</code></td> <td><code>object[]</code></td> <td>Lifecycle actions run after scoring.</td> </tr> <tr> <td><code>agents</code></td> <td><code>string[]</code></td> <td>Override which agents run this scenario.</td> </tr> <tr> <td><code>skills</code></td> <td><code>string[]</code></td> <td>Scenario-specific skills.</td> </tr> </tbody> </table> <div class="callout callout-info"> <div class="callout-title">Setup and Teardown</div> <p>
-Lifecycle actions run sequentially with a 30-second timeout per action. Setup failures abort
-      the job. Teardown failures are logged but do not block subsequent jobs.
-</p> </div> </code> <footer class="site-footer">
-AXIS is maintained by Netlify.
-</footer> </main> </div> <script>
-      const btn = document.getElementById("menu-btn");
-      const sidebar = document.getElementById("sidebar");
-      const backdrop = document.getElementById("sidebar-backdrop");
-
-      function toggle() {
-        sidebar.classList.toggle("open");
-        backdrop.classList.toggle("open");
-      }
-
-      btn.addEventListener("click", toggle);
-      backdrop.addEventListener("click", toggle);
-    </script> </body> </html>

package/dist/docs-site/content-assets.mjs

@@ -1 +0,0 @@
-export default new Map();

package/dist/docs-site/content-modules.mjs

@@ -1 +0,0 @@
-export default new Map();

package/dist/docs-site/data-store.json

@@ -1,9 +0,0 @@
-[
-  ["Map", 1, 2],
-  "meta::meta",
-  ["Map", 3, 4, 5, 6],
-  "astro-version",
-  "5.18.1",
-  "astro-config-digest",
-  "{\"root\":{},\"srcDir\":{},\"publicDir\":{},\"outDir\":{},\"cacheDir\":{},\"compressHTML\":true,\"base\":\"/\",\"trailingSlash\":\"ignore\",\"output\":\"static\",\"scopedStyleStrategy\":\"attribute\",\"build\":{\"format\":\"directory\",\"client\":{},\"server\":{},\"assets\":\"_astro\",\"serverEntry\":\"entry.mjs\",\"redirects\":true,\"inlineStylesheets\":\"auto\",\"concurrency\":1},\"server\":{\"open\":false,\"host\":false,\"port\":4321,\"streaming\":true,\"allowedHosts\":[]},\"redirects\":{},\"image\":{\"endpoint\":{\"route\":\"/_image\"},\"service\":{\"entrypoint\":\"astro/assets/services/sharp\",\"config\":{}},\"domains\":[],\"remotePatterns\":[],\"responsiveStyles\":false},\"devToolbar\":{\"enabled\":true},\"markdown\":{\"syntaxHighlight\":{\"type\":\"shiki\",\"excludeLangs\":[\"math\"]},\"shikiConfig\":{\"langs\":[],\"langAlias\":{},\"theme\":\"github-dark\",\"themes\":{},\"wrap\":false,\"transformers\":[]},\"remarkPlugins\":[],\"rehypePlugins\":[],\"remarkRehype\":{},\"gfm\":true,\"smartypants\":true},\"security\":{\"checkOrigin\":true,\"allowedDomains\":[],\"actionBodySizeLimit\":1048576},\"env\":{\"schema\":{},\"validateSecrets\":false},\"experimental\":{\"clientPrerender\":false,\"contentIntellisense\":false,\"headingIdCompat\":false,\"preserveScriptOrder\":false,\"liveContentCollections\":false,\"csp\":false,\"staticImportMetaEnv\":false,\"chromeDevtoolsWorkspace\":false,\"failOnPrerenderConflict\":false,\"svgo\":false},\"legacy\":{\"collections\":false}}"
-]

package/dist/docs-site/index.html

@@ -1,69 +0,0 @@
-<!DOCTYPE html><html lang="en"> <head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>What is AXIS - AXIS Docs</title><meta name="description" content="Documentation for AXIS, the Agent eXperience Index Score synthetic testing framework for AI agents."><link rel="stylesheet" href="/_astro/cli.DDWZtG0-.css"></head> <body> <header class="site-header"> <a href="/" class="site-logo" aria-label="AXIS home"> <span class="site-logo-mark"><span class="logo-ax">AX</span><span class="logo-i">I</span>S</span> </a> <div class="site-header-links"> <a href="https://github.com/netlify/axis">GitHub</a> <a href="https://www.npmjs.com/package/@netlify/axis">npm</a> </div> <button class="mobile-menu-btn" id="menu-btn" aria-label="Toggle navigation">&#9776;</button> </header> <div class="site-shell"> <div class="sidebar-backdrop" id="sidebar-backdrop"></div> <aside class="sidebar" id="sidebar"> <nav class="sidebar-nav"> <div class="sidebar-section-label">Getting Started</div> <a href="/" class="sidebar-link active">What is AXIS</a> <a href="/quickstart" class="sidebar-link">Quick Start</a> <div class="sidebar-section-label">How It Works</div> <a href="/scoring" class="sidebar-link">Scoring Framework</a> <a href="/running" class="sidebar-link">Running Tests</a> <div class="sidebar-section-label">Reference</div> <a href="/cli" class="sidebar-link">CLI</a> <a href="/configuration" class="sidebar-link">Configuration</a> </nav> </aside> <main class="main-content">  <div class="hero"> <h1>AXIS - Agent eXperience Index Score</h1> <p class="lead">
-AXIS is both an open scoring framework for measuring <strong>agent experience (AX)</strong>
-and a CLI tool that implements it. Think
-<a href="https://developer.chrome.com/docs/lighthouse">Lighthouse</a>, but instead of scoring
-      user experience, AXIS scores <strong>agent experience</strong>.
-</p> </div> <h2>Why AX Matters</h2> <p>
-The web has Lighthouse. APIs have contract testing. Performance has k6. But there is no
-    standardized way to answer: <em>"How well does my system work when an AI agent tries to use it?"</em> </p> <p>
-As agents become a primary interface for websites, APIs, and developer platforms, the systems
-    they interact with need to be measured and optimized for that interaction. Just as we optimize
-    for page load time or accessibility, AX is the agent-era equivalent of UX.
-</p> <h2>Our Approach</h2> <p>
-AXIS is built on two core beliefs about how agent experience should be measured.
-</p> <p> <strong>Measure what matters, where you have leverage.</strong> Agent experience is not a single
-    number. It breaks down into distinct dimensions: how well the agent completes the task, how it
-    uses the environment, how it interacts with services, and how it reasons through problems.
-    Purpose-built tooling that scores each dimension independently gives you a clear picture of
-    where to focus. If your API responses are slowing agents down, you see it in the Service
-    dimension. If your project structure confuses agents, Environment tells you. Generic pass/fail
-    testing does not surface these signals.
-</p> <p> <strong>Test against real agent behavior, not theoretical support.</strong> It is not enough to
-    validate that your system publishes the right config files or follows a protocol spec. What
-    matters is whether agents actually discover and use what you provide. AXIS measures this by
-    running real agents against real scenarios and observing what happens: which tools they call,
-    which files they read, which APIs they hit. This tells you what agents <em>do</em>, not what
-    they <em>could</em> do in theory.
-</p> <h2>The Scoring Framework</h2> <p>
-At its core, AXIS defines a standard way to measure agent experience across four independent
-    dimensions. Any tool, platform, or CI system can implement this framework to produce
-    comparable AX measurements.
-</p> <div class="card-grid"> <div class="card card-accent"> <span class="card-weight">40%</span> <div class="card-title">Goal Achievement</div> <div class="card-desc">
-Did the agent complete the task? Evaluated against rubric criteria you define for each
-        scenario.
-</div> </div> <div class="card card-env"> <span class="card-weight">20%</span> <div class="card-title">Environment</div> <div class="card-desc">
-How well did the agent use the OS, filesystem, and dev tools? Measures quality of shell
-        commands, file operations, git usage, and build tools.
-</div> </div> <div class="card card-svc"> <span class="card-weight">20%</span> <div class="card-title">Service</div> <div class="card-desc">
-How effectively did the agent use external services? Evaluates API calls, MCP tools,
-        network requests, and third-party integrations.
-</div> </div> <div class="card card-agent"> <span class="card-weight">20%</span> <div class="card-title">Agent</div> <div class="card-desc">
-How well did the agent reason and self-organize? Covers planning, task management, tool
-        discovery, and metacognitive behavior.
-</div> </div> </div> <p>
-These four dimensions combine into a single 0 to 100 <strong>AXIS Result</strong>. The
-    framework specifies what signals feed each dimension, how interactions are categorized, and
-    how the composite score is calculated. See <a href="/scoring">Scoring Framework</a> for full
-    details on the signals and scoring logic.
-</p> <h2>The CLI Tool</h2> <p>
-The <code>@netlify/axis</code> package is the reference implementation of the scoring framework.
-    It provides a CLI that runs agent scenarios, captures transcripts, scores the results, and
-    produces reports.
-</p> <ul> <li><strong>Define scenarios</strong> as JSON files with a prompt, rubric, and optional setup/teardown steps.</li> <li><strong>Run them</strong> against any supported agent (or your own custom adapter) in isolated workspaces.</li> <li><strong>Score automatically</strong> using a multi-pass LLM evaluation pipeline that produces per-dimension and composite scores.</li> <li><strong>Track over time</strong> with persistent reports, baseline snapshots, and regression detection for CI gating.</li> </ul> <h3>Built-in Adapters</h3> <p>
-The CLI ships with adapters for popular AI coding agents. Each adapter handles process spawning,
-    transcript capture, and output normalization for its agent.
-</p> <table> <thead> <tr> <th>Adapter</th> <th>Agent</th> <th>Required Env</th> </tr> </thead> <tbody> <tr> <td><code>claude-code</code></td> <td>Claude Code</td> <td><code>ANTHROPIC_API_KEY</code></td> </tr> <tr> <td><code>codex</code></td> <td>OpenAI Codex</td> <td><code>CODEX_API_KEY</code></td> </tr> <tr> <td><code>gemini</code></td> <td>Google Gemini CLI</td> <td><code>GEMINI_API_KEY</code></td> </tr> <tr> <td><code>goose</code></td> <td>Goose</td> <td>None</td> </tr> <tr> <td><code>claude-sdk</code></td> <td>Claude SDK</td> <td><code>ANTHROPIC_API_KEY</code></td> </tr> <tr> <td><code>gemini-acp</code></td> <td>Gemini (ACP)</td> <td><code>GEMINI_API_KEY</code></td> </tr> <tr> <td>Custom</td> <td>Any agent via <code>createAgentAdapter()</code></td> <td>User-defined</td> </tr> </tbody> </table>  <footer class="site-footer">
-AXIS is maintained by Netlify.
-</footer> </main> </div> <script>
-      const btn = document.getElementById("menu-btn");
-      const sidebar = document.getElementById("sidebar");
-      const backdrop = document.getElementById("sidebar-backdrop");
-
-      function toggle() {
-        sidebar.classList.toggle("open");
-        backdrop.classList.toggle("open");
-      }
-
-      btn.addEventListener("click", toggle);
-      backdrop.addEventListener("click", toggle);
-    </script> </body> </html>

package/dist/docs-site/quickstart/index.html

@@ -1,59 +0,0 @@
-<!DOCTYPE html><html lang="en"> <head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>Quick Start - AXIS Docs</title><meta name="description" content="Documentation for AXIS, the Agent eXperience Index Score synthetic testing framework for AI agents."><link rel="stylesheet" href="/_astro/cli.DDWZtG0-.css"></head> <body> <header class="site-header"> <a href="/" class="site-logo" aria-label="AXIS home"> <span class="site-logo-mark"><span class="logo-ax">AX</span><span class="logo-i">I</span>S</span> </a> <div class="site-header-links"> <a href="https://github.com/netlify/axis">GitHub</a> <a href="https://www.npmjs.com/package/@netlify/axis">npm</a> </div> <button class="mobile-menu-btn" id="menu-btn" aria-label="Toggle navigation">&#9776;</button> </header> <div class="site-shell"> <div class="sidebar-backdrop" id="sidebar-backdrop"></div> <aside class="sidebar" id="sidebar"> <nav class="sidebar-nav"> <div class="sidebar-section-label">Getting Started</div> <a href="/" class="sidebar-link">What is AXIS</a> <a href="/quickstart" class="sidebar-link active">Quick Start</a> <div class="sidebar-section-label">How It Works</div> <a href="/scoring" class="sidebar-link">Scoring Framework</a> <a href="/running" class="sidebar-link">Running Tests</a> <div class="sidebar-section-label">Reference</div> <a href="/cli" class="sidebar-link">CLI</a> <a href="/configuration" class="sidebar-link">Configuration</a> </nav> </aside> <main class="main-content">  <h1>Quick Start</h1> <p class="lead">
-Get AXIS running in your project in a few minutes. This guide walks through creating a config,
-    writing your first scenario, and viewing the results.
-</p> <h2>Prerequisites</h2> <ul> <li>Node.js 18 or later.</li> <li>An API key for at least one supported agent (for example, <code>ANTHROPIC_API_KEY</code> for Claude Code).</li> </ul> <h2>1. Create a Config File</h2> <p>
-Add an <code>axis.config.json</code> to your project root. At minimum, specify where your
-    scenarios live and which agents to run.
-</p> <pre><code>{
-  &quot;scenarios&quot;: &quot;./scenarios&quot;,
-  &quot;agents&quot;: [&quot;claude-code&quot;]
-}</code></pre> <h2>2. Write a Scenario</h2> <p>
-Create a <code>scenarios/</code> directory and add your first scenario as a JSON file. Each
-    scenario needs a name, a prompt (the task for the agent), and a rubric (your success criteria).
-</p> <pre><code>{
-  &quot;name&quot;: &quot;Create a greeting file&quot;,
-  &quot;prompt&quot;: &quot;Create a file called hello.txt with the content &#39;Hello from AXIS&#39;.&quot;,
-  &quot;rubric&quot;: [
-    { &quot;check&quot;: &quot;File hello.txt exists&quot;, &quot;weight&quot;: 0.5 },
-    { &quot;check&quot;: &quot;File contains &#39;Hello from AXIS&#39;&quot;, &quot;weight&quot;: 0.5 }
-  ]
-}</code></pre> <p>
-Save this as <code>scenarios/hello-world.json</code>. The filename (without <code>.json</code>)
-    becomes the scenario key used in reports and CLI commands.
-</p> <h2>3. Run It</h2> <pre><code>npx @netlify/axis run</code></pre> <p>
-AXIS spawns the agent in an isolated workspace, captures the full interaction transcript, scores
-    the result against your rubric, and displays a report in your terminal.
-</p> <h2>4. View the Report</h2> <p>
-Every run saves a report to <code>.axis/reports/</code>. You can view it again at any time.
-</p> <pre><code># View the latest report summary
-npx @netlify/axis reports latest
-
-# Open the HTML report in your browser
-npx @netlify/axis reports latest --html
-
-# Get JSON output for scripting
-npx @netlify/axis reports latest --json</code></pre> <h2>5. Set a Baseline</h2> <p>
-Once you have a run you are happy with, save it as a baseline. Future runs can diff against it
-    to detect regressions.
-</p> <pre><code># Save the latest report as a baseline
-npx @netlify/axis baseline set
-
-# Compare future runs automatically
-npx @netlify/axis run --compare-baseline</code></pre> <p>
-The diff command exits with code 1 if any regressions are detected, making it suitable for CI
-    gating.
-</p> <h2>Next Steps</h2> <ul> <li><a href="/scoring">Scoring Framework</a> explains how the four dimensions are calculated and what signals drive each score.</li> <li><a href="/running">Running Tests</a> covers the full config reference, CLI commands, custom adapters, MCP servers, and CI integration.</li> </ul>  <footer class="site-footer">
-AXIS is maintained by Netlify.
-</footer> </main> </div> <script>
-      const btn = document.getElementById("menu-btn");
-      const sidebar = document.getElementById("sidebar");
-      const backdrop = document.getElementById("sidebar-backdrop");
-
-      function toggle() {
-        sidebar.classList.toggle("open");
-        backdrop.classList.toggle("open");
-      }
-
-      btn.addEventListener("click", toggle);
-      backdrop.addEventListener("click", toggle);
-    </script> </body> </html>

package/dist/docs-site/running/index.html

@@ -1,87 +0,0 @@
-<!DOCTYPE html><html lang="en"> <head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>Running Tests - AXIS Docs</title><meta name="description" content="Documentation for AXIS, the Agent eXperience Index Score synthetic testing framework for AI agents."><link rel="stylesheet" href="/_astro/cli.DDWZtG0-.css"></head> <body> <header class="site-header"> <a href="/" class="site-logo" aria-label="AXIS home"> <span class="site-logo-mark"><span class="logo-ax">AX</span><span class="logo-i">I</span>S</span> </a> <div class="site-header-links"> <a href="https://github.com/netlify/axis">GitHub</a> <a href="https://www.npmjs.com/package/@netlify/axis">npm</a> </div> <button class="mobile-menu-btn" id="menu-btn" aria-label="Toggle navigation">&#9776;</button> </header> <div class="site-shell"> <div class="sidebar-backdrop" id="sidebar-backdrop"></div> <aside class="sidebar" id="sidebar"> <nav class="sidebar-nav"> <div class="sidebar-section-label">Getting Started</div> <a href="/" class="sidebar-link">What is AXIS</a> <a href="/quickstart" class="sidebar-link">Quick Start</a> <div class="sidebar-section-label">How It Works</div> <a href="/scoring" class="sidebar-link">Scoring Framework</a> <a href="/running" class="sidebar-link active">Running Tests</a> <div class="sidebar-section-label">Reference</div> <a href="/cli" class="sidebar-link">CLI</a> <a href="/configuration" class="sidebar-link">Configuration</a> </nav> </aside> <main class="main-content">  <h1>Running AXIS Tests</h1> <p class="lead">
-How AXIS executes scenarios, manages agent processes, and produces reports.
-</p> <h2>Execution Model</h2> <p>
-When you run <code>axis run</code>, AXIS loads your config, discovers scenarios, and executes
-    each scenario/agent combination as an independent job. Jobs run in parallel up to the configured
-    concurrency limit.
-</p> <p>
-Each job follows the same lifecycle:
-</p> <ol> <li>Run setup actions (if defined in the scenario).</li> <li>Spawn the agent process in an isolated workspace.</li> <li>Stream and capture the full interaction transcript.</li> <li>Score the transcript against the rubric (unless <code>--no-score</code> is set).</li> <li>Run teardown actions (if defined).</li> <li>Save the result to the report.</li> </ol> <h2>Built-in Adapters</h2> <p>
-AXIS ships with adapters for popular AI coding agents. Each adapter handles CLI resolution,
-    process spawning, transcript capture, and output normalization.
-</p> <table> <thead> <tr> <th>Adapter</th> <th>CLI Binary</th> <th>Required Env</th> <th>Default Flags</th> </tr> </thead> <tbody> <tr> <td><code>claude-code</code></td> <td><code>claude</code></td> <td><code>ANTHROPIC_API_KEY</code></td> <td><code>dangerously-skip-permissions</code></td> </tr> <tr> <td><code>codex</code></td> <td><code>codex</code></td> <td><code>CODEX_API_KEY</code></td> <td><code>full-auto</code>, <code>skip-git-repo-check</code></td> </tr> <tr> <td><code>gemini</code></td> <td><code>gemini</code></td> <td><code>GEMINI_API_KEY</code></td> <td><code>yolo</code></td> </tr> <tr> <td><code>goose</code></td> <td><code>goose</code></td> <td>None</td> <td>None</td> </tr> <tr> <td><code>claude-sdk</code></td> <td>SDK</td> <td><code>ANTHROPIC_API_KEY</code></td> <td>None</td> </tr> <tr> <td><code>gemini-acp</code></td> <td>ACP</td> <td><code>GEMINI_API_KEY</code></td> <td>None</td> </tr> </tbody> </table> <p>
-CLI binaries are resolved automatically. If not found locally, AXIS falls back to
-<code>npx --yes &lt;package&gt;</code> silently.
-</p> <h2>Custom Adapters</h2> <p>
-Create a custom adapter module using <code>createAgentAdapter()</code> and register it in your
-<a href="/configuration">config</a>.
-</p> <pre><code>// adapters/my-agent.ts
-import { createAgentAdapter } from &quot;@netlify/axis&quot;;
-
-export default createAgentAdapter&lt;{ stdout: string }&gt;({
-  name: &quot;my-agent&quot;,
-  resolveCommand: () =&gt; ({ command: &quot;my-cli&quot;, prefixArgs: [] }),
-  buildArgs: (input) =&gt; [input.prompt],
-  initialState: () =&gt; ({ stdout: &quot;&quot; }),
-  streamConfig: {
-    mode: &quot;aggregate&quot;,
-    onChunk: (chunk, ctx) =&gt; {
-      ctx.state.stdout += chunk;
-    },
-  },
-  getResult: (ctx) =&gt; ({
-    result: ctx.state.stdout.trim() || null,
-  }),
-});</code></pre> <p>
-Adapters support two stream modes: <strong>lines</strong> (NDJSON, one JSON object per stdout
-    line) and <strong>aggregate</strong> (raw chunks accumulated in state). The module must export
-    an <code>AgentAdapter</code> as default or as a named <code>adapter</code> export.
-</p> <h2>Workspace Isolation</h2> <p>
-Each agent run gets a fresh temporary directory as its workspace. AXIS isolates the following.
-</p> <ul> <li><strong>HOME directory:</strong> Set to the workspace to prevent global config leakage.</li> <li><strong>Adapter-specific dirs:</strong> <code>CLAUDE_CONFIG_DIR</code>, <code>CODEX_HOME</code>, <code>GEMINI_CLI_HOME</code>.</li> <li><strong>Environment variables:</strong> Only explicitly listed vars and system essentials are passed through.</li> </ul> <h2>Reports</h2> <p>
-Every run automatically saves a report to <code>.axis/reports/</code>.
-</p> <pre><code>.axis/reports/{reportId}/
-  report.json                          # Manifest with summary + metadata
-  scenarios/{key}/{agent}.json         # Full result with transcript
-  scenarios/{key}/{agent}.raw.ndjson   # Raw stdout (--debug only)
-  scenarios/{key}/{agent}.sparse-index.txt  # Scoring reference</code></pre> <p>
-Use <code>axis reports</code> to list, view, and export reports. See the
-<a href="/cli">CLI reference</a> for all available options.
-</p> <h2>Baselines</h2> <p>
-Baselines let you snapshot scores and detect regressions in future runs. They are stored in
-<code>.axis/baselines/</code> and designed to be checked into version control.
-</p> <pre><code># 1. Run your scenarios
-axis run
-
-# 2. Save the results as a baseline
-axis baseline set
-
-# 3. In future runs, compare against the baseline
-axis run --compare-baseline
-
-# 4. Or diff explicitly
-axis baseline diff</code></pre> <p>
-Baseline diff uses a noise tolerance of 1 point. Score deltas of 0 to 1 are reported
-    as unchanged. The diff command exits with code 1 if any regressions are detected, making it
-    suitable for CI gating.
-</p> <h2>CI Integration</h2> <p>
-AXIS is designed to work in CI environments. Key patterns:
-</p> <ul> <li>Use <code>--json</code> for machine-readable output.</li> <li>Use <code>--compare-baseline</code> to gate on regressions (exit code 1).</li> <li>Set <code>--concurrency</code> to control resource usage.</li> <li>Pass API keys via environment variables.</li> </ul> <pre><code># GitHub Actions example
-- name: Run AXIS tests
-  env:
-    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
-  run: npx @netlify/axis run --json --compare-baseline</code></pre>  <footer class="site-footer">
-AXIS is maintained by Netlify.
-</footer> </main> </div> <script>
-      const btn = document.getElementById("menu-btn");
-      const sidebar = document.getElementById("sidebar");
-      const backdrop = document.getElementById("sidebar-backdrop");
-
-      function toggle() {
-        sidebar.classList.toggle("open");
-        backdrop.classList.toggle("open");
-      }
-
-      btn.addEventListener("click", toggle);
-      backdrop.addEventListener("click", toggle);
-    </script> </body> </html>