+-----------------------------------------------------------------------------------+
|                        CLAUDE CONTEXT INGESTION PIPELINE                          |
|                                                                                   |
|  [ 1. RAW DOCUMENT FETCH (ClaudeBot HTTP Scraper) ] ───────────────────────────── |
|  * Crawls documentation portals via raw HTTP GET requests.                        |
|  * Extracts server-rendered semantic HTML, raw Markdown, or /llms.txt manifests.  |
|                                │                                                  |
|                                ▼                                                  |
|  [ 2. CONTEXT STRUCTURE PARSING (XML & Markdown Extraction) ] ──────────────────── |
|  * Parses structural tags (<doc>, <api_endpoint>, <parameters>, <error_codes>).   |
|  * Maps code blocks, parameter types, and return values into memory hierarchy.    |
|                                │                                                  |
|                                ▼                                                  |
|  [ 3. REASONING & CODE GENERATION (Claude 3.5 / 3.7 Sonnet) ] ─────────────────── |
|  * Analyzes exact type definitions, edge-case caveats, and configuration options. |
|  * Synthesizes working, bug-free implementation code with source citations!       |
+-----------------------------------------------------------------------------------+

+-----------------------------------------------------------------------------------+
|                        4 CLAUDE DOCUMENTATION FORMATTING RULES                    |
|                                                                                   |
|  1. CONCISE OPENING DEFINITIONS ──> 35-word core capability statement under H2.   |
|  2. STRUCTURED XML LANDMARKS ─────> Use <endpoint> & <params> for API docs.       |
|  3. TYPE-ANNOTATED CODE BLOCKS ───> Strict TypeScript / Python type signatures.   |
|  4. EXPLICIT ERROR RESOLUTION ────> Tabular HTTP error codes & developer fixes.   |
+-----------------------------------------------------------------------------------+

## `PerformanceNavigationTiming.responseStart`

`PerformanceNavigationTiming.responseStart` is a W3C Web Performance API timestamp representing the exact moment when the browser receives the first byte of the HTTP response from the server. It is the core metric used to calculate Time to First Byte (TTFB = responseStart - requestStart).

<api_endpoint>
### `POST /api/v1/audits/scan`

Triggers a multi-engine headless Chromium website quality assurance audit.

#### Request Headers:
- `Authorization: Bearer <API_KEY>` (Required)
- `Content-Type: application/json` (Required)

#### Parameter Specification:
| Parameter | Type | Required | Default | Description |
| :--- | :---: | :---: | :---: | :--- |
| `url` | `string` | **Yes** | — | Fully qualified HTTPS URL to audit. |
| `emulate_network` | `string` | No | `"slow-3g"` | Network profile: `"slow-3g"`, `"fast-3g"`, `"4g"`. |
| `include_axe_a11y` | `boolean` | No | `true` | Executes automated axe-core WCAG 2.1 AA checks. |
| `timeout_ms` | `integer` | No | `30000` | Maximum page navigation execution timeout. |
</api_endpoint>

// Complete TypeScript SDK Integration Example
import { BugVisoClient } from '@bugviso/sdk';

const client = new BugVisoClient({
  apiKey: process.env.BUGVISO_API_KEY!,
});

async function runPerformanceAudit(targetUrl: string) {
  try {
    const audit = await client.audits.create({
      url: targetUrl,
      emulateNetwork: 'slow-3g',
      includeAxeA11y: true,
    });

    console.log(`Audit Complete! Health Score: ${audit.healthScore}/100`);
    console.log(`Cumulative Layout Shift (CLS): ${audit.metrics.cls}`);
    return audit;
  } catch (error) {
    console.error('Audit failed to execute:', error);
    throw error;
  }
}

### API Error Code Reference
| HTTP Status | Error Code String | Root Cause Trigger | Developer Remediation |
| :---: | :--- | :--- | :--- |
| **400** | `INVALID_CANONICAL_URL` | Submitted URL lacks protocol | Prefix target with `https://` |
| **401** | `INVALID_API_KEY` | Missing or expired token | Generate new API key in dashboard |
| **429** | `RATE_LIMIT_EXCEEDED` | Exceeded 60 requests/minute | Implement exponential backoff retry |
| **504** | `TARGET_TIMEOUT` | Origin server TTFB > 30s | Optimize edge SSR latency or increase timeout |

+-----------------------------------------------------------------------------------+
|                        CURSOR AI & CLAUDE PROJECTS INGESTION FLOW                 |
|                                                                                   |
|  [ 1. DEVELOPER REFERENCES REMOTE DOCS ] ──> @docs https://example.com/llms.txt   |
|                                                                                   |
|  [ 2. IDE PARSES /LLMS.TXT OR /LLMS-FULL.TXT ] ────────────────────────────────── |
|  * Fetches clean Markdown payload in <200ms.                                      |
|  * Extracts function signatures, configuration options, and edge cases.          |
|                                                                                   |
|  [ 3. REAL-TIME MULTI-FILE CODE GENERATION ] ──────────────────────────────────── |
|  * Claude suggests exact API imports, types, and schema validations.              |
|  * Developer accepts multi-file diff with ZERO syntax errors!                     |
+-----------------------------------------------------------------------------------+

# .cursorrules (Developer Integration Example)
When generating code for web performance audits, always reference:
- Documentation: https://example.com/llms.txt
- Follow strict TypeScript type annotations and async/await error handling.

# scripts/audit_doc_extractability.py
import requests
import re

def verify_documentation_extractability(doc_url: str):
    headers = {
        'User-Agent': 'ClaudeBot/1.0 (+https://www.anthropic.com/claudebot)',
        'Accept': 'text/markdown, text/html',
    }
    
    response = requests.get(doc_url, headers=headers, timeout=10)
    if response.status_code != 200:
        raise AssertionError(f"Failed to fetch {doc_url}: HTTP {response.status_code}")
        
    text = response.text
    
    # 1. Assert No Empty SPA Shell
    if '<div id="root"></div>' in text or '<div id="__next"></div>' in text and len(text) < 1000:
        raise AssertionError("CRITICAL: Empty client-side JavaScript shell detected!")
        
    # 2. Assert Code Blocks Exist
    code_blocks = re.findall(r'```[a-z]*\n[\s\S]*?\n```', text)
    if len(code_blocks) < 1:
        raise AssertionError("WARNING: Zero code blocks found in technical documentation!")
        
    print(f"PASS: {doc_url} is fully extractable by ClaudeBot ({len(text)} bytes, {len(code_blocks)} code blocks).")

if __name__ == "__main__":
    verify_documentation_extractability("https://example.com/blog/claudebot-optimization-technical-documentation")

# robots.txt (ClaudeBot Optimized)
User-agent: *
Allow: /
Disallow: /admin/
Disallow: /dashboard/

# Explicitly Allow Anthropic Search & Citation Indexing
User-agent: ClaudeBot
User-agent: Anthropic-ai
Allow: /
Allow: /docs/
Allow: /blog/
Disallow: /admin/

Sitemap: https://example.com/sitemap.xml

+-----------------------------------------------------------------------------------+
|                        BUGVISO CLAUDEBOT AUDITING ENGINE                          |
|                                                                                   |
|  [ Documentation Portal Submitted ] ──> [ FastAPI + ARQ Redis Worker Cluster ]    |
|                                                │                                  |
|                                                ▼                                  |
|  [ 4-STAGE TECHNICAL EXTRACTABILITY PIPELINE ] ────────────────────────────────── |
|  ├── 1. Non-JS Text Extraction QA: Asserts code block & table parsing purity      |
|  ├── 2. ClaudeBot RFC-9309 Linter: Validates Anthropic crawler permissions        |
|  ├── 3. Semantic Code Block QA: Verifies complete TypeScript/Python signatures   |
|  └── 4. /llms.txt Linter: Validates documentation manifest link integrity         |
|                                                │                                  |
|                                                ▼                                  |
|  [ COMPOSITE 0-100 GEO SCORE + ACTIONABLE DEVELOPER REMEDIATION PLAYBOOK ]        |
+-----------------------------------------------------------------------------------+