The llms.txt Audit Template: Score AI Readiness in 10 Mins
Download the master llms.txt audit template score framework in 2026. Evaluate your /llms.txt syntax, markdown headers, and AI crawler links in 10 minutes.
The llms.txt Audit Template: Score AI Readiness in 10 Mins
In 2026, the /llms.txt standard has emerged as one of the highest-leverage technical assets for web applications, SaaS documentation portals, and enterprise knowledge bases seeking visibility in generative artificial intelligence search engines. Functioning as a curated, machine-readable Markdown index for large language models, /llms.txt allows AI search crawlers (including ChatGPT Search, Claude, and Perplexity) and developer tooling (such as Cursor IDE) to ingest your platform's core capabilities without parsing layout boilerplate or traversing complex HTML DOM trees.
However, simply publishing an arbitrary text file at /llms.txt does not guarantee AI citability. If your manifest contains broken markdown links, missing summary blockquotes, non-standard MIME headers, or contradicts your robots.txt crawler permissions, AI retrieval engines will reject the file during ingestion. To achieve peak Generative Engine Optimization (GEO) performance, engineering and marketing teams need a structured scoring framework to audit their /llms.txt deployment against formal industry standards.
In this deep-dive technical template guide, we present the llms.txt audit template score framework. We break down the 15-point scoring methodology across four critical dimensions (Syntax & Formatting, Content Depth & Structure, Technical Infrastructure & MIME, and Target Link Health), provide an automated Python scoring script, offer a downloadable evaluation spreadsheet model, and demonstrate how to audit your implementation using modern cloud diagnostics.
The 100-Point /llms.txt Readiness Scoring Model
Our audit framework evaluates /llms.txt implementations on a standardized 0–100 percentage scale across four weighted categories:
+-----------------------------------------------------------------------------------+
| THE 100-POINT /LLMS.TXT SCORING WEIGHTS |
| |
| [ 1. SYNTAX & SPECIFICATION COMPLIANCE (30 Points) ] ─────────────────────────── |
| * Root URL resolution, H1 title, summary blockquote, markdown bullet syntax. |
| │ |
| ▼ |
| [ 2. INFRASTRUCTURE & MIME HEADERS (25 Points) ] ─────────────────────────────── |
| * text/plain Content-Type, UTF-8 charset, edge caching, sub-50ms TTFB. |
| │ |
| ▼ |
| [ 3. TARGET LINK HEALTH & INTEGRITY (25 Points) ] ────────────────────────────── |
| * 100% HTTP 200 OK links, absolute HTTPS URLs, zero redirect chains. |
| │ |
| ▼ |
| [ 4. CORPUS COMPLETENESS & /LLMS-FULL.TXT (20 Points) ] ──────────────────────── |
| * Presence of /llms-full.txt inlined corpus, optional section, freshness sync. |
+-----------------------------------------------------------------------------------+The 15-Point /llms.txt Master Audit Matrix
Evaluate your website's /llms.txt file against this 15-point scoring matrix:
| # | Dimension | Critical Audit Check | Max Points | Pass Criteria |
|---|---|---|---|---|
| 1 | Syntax | Root URL Resolution | 10 pts | File resolves at https://example.com/llms.txt |
| 2 | Syntax | H1 Project Title | 5 pts | Exactly one single # Project Name header |
| 3 | Syntax | Summary Blockquote | 10 pts | Direct > 50-word capability summary beneath H1 |
| 4 | Syntax | Markdown Bullet Links | 5 pts | Clean - [Title](URL): Description format |
| 5 | Infra | Content-Type Header | 10 pts | Content-Type: text/plain; charset=utf-8 |
| 6 | Infra | Sub-50ms Edge TTFB | 5 pts | Response served from Edge CDN network |
| 7 | Infra | Cache-Control Header | 5 pts | s-maxage=3600, stale-while-revalidate set |
| 8 | Infra | Robots Exclusion Access | 5 pts | robots.txt grants AI bots read access |
| 9 | Links | Absolute HTTPS URLs | 10 pts | All links prefixed with https://domain.com |
| 10 | Links | Zero Broken Links (404) | 10 pts | 100% of declared markdown links return HTTP 200 |
| 11 | Links | Zero Redirect Chains | 5 pts | Links point directly to canonical destinations |
| 12 | Corpus | /llms-full.txt Exists | 10 pts | Complete inlined corpus available at root |
| 13 | Corpus | Optional Section Link | 5 pts | ## Optional section links to /llms-full.txt |
| 14 | Corpus | Semantic H2 Grouping | 3 pts | Links grouped into logical H2 categories |
| 15 | Corpus | Freshness Synchronization | 2 pts | Synchronized with latest CMS content |
| TOTAL | — | COMPOSITE GEO SCORE | 100 pts | Target: ≥ 90 Points for AI Citability |
Automated Python CLI Script to Calculate /llms.txt Score
Execute this automated auditing script to score your /llms.txt file in under 10 seconds:
# scripts/audit_llms_txt_score.py
import requests
import re
def calculate_llms_txt_score(domain_url: str):
score = 0
breakdown = {}
target_url = f"{domain_url.rstrip('/')}/llms.txt"
try:
r = requests.get(target_url, timeout=10)
# 1. Root URL Resolution (10 pts)
if r.status_code == 200:
score += 10
breakdown["1. Root URL 200 OK"] = "10/10"
else:
breakdown["1. Root URL 200 OK"] = f"0/10 (HTTP {r.status_code})"
return score, breakdown
content = r.text
# 2. H1 Project Title (5 pts)
h1_match = re.search(r'^#\s+(.+)#x27;, content, re.MULTILINE)
if h1_match:
score += 5
breakdown["2. H1 Title Present"] = "5/5"
else:
breakdown["2. H1 Title Present"] = "0/5"
# 3. Summary Blockquote (10 pts)
quote_match = re.search(r'^>\s+(.+)#x27;, content, re.MULTILINE)
if quote_match:
score += 10
breakdown["3. Summary Blockquote"] = "10/10"
else:
breakdown["3. Summary Blockquote"] = "0/10"
# 4. Content-Type Header (10 pts)
ct = r.headers.get("Content-Type", "")
if "text/plain" in ct:
score += 10
breakdown["4. text/plain MIME"] = "10/10"
else:
breakdown["4. text/plain MIME"] = f"0/10 ({ct})"
# 5. Markdown Bullet Links & Target Health (25 pts)
links = re.findall(r'\[([^\]]+)\]\((https?://[^\)]+)\)', content)
if links:
score += 10
breakdown["5. Absolute HTTPS Links"] = "10/10"
# Test sample of up to 5 links
dead_links = 0
for title, url in links[:5]:
try:
head_r = requests.head(url, timeout=5, allow_redirects=True)
if head_r.status_code != 200:
dead_links += 1
except Exception:
dead_links += 1
if dead_links == 0:
score += 15
breakdown["6. Link Health (Sample)"] = "15/15"
else:
breakdown["6. Link Health (Sample)"] = f"0/15 ({dead_links} failed)"
else:
breakdown["5. Absolute HTTPS Links"] = "0/10"
breakdown["6. Link Health (Sample)"] = "0/15"
# 6. Check for /llms-full.txt (20 pts)
full_url = f"{domain_url.rstrip('/')}/llms-full.txt"
full_r = requests.get(full_url, timeout=5)
if full_r.status_code == 200 and len(full_r.text) > 500:
score += 20
breakdown["7. /llms-full.txt Corpus"] = "20/20"
else:
breakdown["7. /llms-full.txt Corpus"] = "0/20"
# 7. Edge TTFB Latency (10 pts)
if r.elapsed.total_seconds() < 0.1:
score += 10
breakdown["8. Sub-100ms TTFB"] = "10/10"
else:
score += 5
breakdown["8. Sub-100ms TTFB"] = f"5/10 ({r.elapsed.total_seconds()*1000:.0f}ms)"
except Exception as e:
breakdown["Fatal Exception"] = str(e)
print(f"Audit Complete for {domain_url}: Total Score = {score}/100")
for k, v in breakdown.items():
print(f" * {k}: {v}")
return score, breakdown
if __name__ == "__main__":
calculate_llms_txt_score("https://bugviso.com")Production /llms.txt Reference Example (100/100 Score)
Below is a certified 100/100 reference implementation adhering to all syntax specifications:
# BugViso Platform AI Documentation
> BugViso is a modern website quality assurance and Generative Engine Optimization (GEO) audit platform. It runs headless Chromium performance scans under 3G throttling, tests automated WCAG 2.1 AA accessibility, and scores machine citability for AI search engines.
## Core Documentation & Product Capabilities
- [Website Quality Assurance Overview](https://bugviso.com/blog/what-is-a-website-audit-a-complete-2026-guide): Complete guide to modern multi-engine web performance and SEO auditing.
- [Generative Engine Optimization Framework](https://bugviso.com/blog/generative-engine-optimization-framework-2026): The 5 architectural pillars of AI search citability.
- [Robots.txt AI Bot Governance Guide](https://bugviso.com/blog/robots-txt-ai-bots-audit-gptbot-claudebot): Configuring RFC-9309 rules for GPTBot, ClaudeBot, and PerplexityBot.
## Technical Knowledge Base & Tutorials
- [Next.js 15 App Router SEO Benchmarks](https://bugviso.com/blog/nextjs-app-router-vs-pages-router-seo-benchmarks): Performance benchmarks across 20 production routes.
- [React Hydration CLS Remediation](https://bugviso.com/blog/react-hydration-layout-shift-cls-fix): Eliminating visual layout shifts during client component hydration.
## Optional
- [Complete Inlined Corpus](https://bugviso.com/llms-full.txt): Complete markdown text of all documentation for deep-context IDE and Claude Projects ingestion.To explore how /llms.txt integrates into broader search architecture, review our guides on the llms txt standard guide syntax, how to create llms txt template guide, and what is llms txt ai website guide.
The Master 10-Point /llms.txt Pre-Deployment Matrix
Before deploying changes, verify your endpoint against this checklist:
| Verification Dimension | Critical Audit Check | Technical Implementation Method | Success Criteria |
|---|---|---|---|
| URL Path | Root Domain Path | Serve at https://example.com/llms.txt | Returns HTTP 200 OK with zero redirects |
| MIME Type | Plain Text Stream | Content-Type: text/plain; charset=utf-8 | Raw text stream with zero HTML wrappers |
| H1 Title & Summary | Project Title & Summary | Single # Title + > Summary blockquote | Concise 50-word project capability overview |
| Markdown Bullets | Valid Link Syntax | - [Title](URL): Description format | Clean link format matching standard specs |
| Absolute URLs | Fully Qualified HTTPS | Prefix all links with https://domain.com | AI models resolve direct targets without error |
| Full Corpus Link | /llms-full.txt Reference | Optional section bullet linking full corpus | Allows full IDE context ingestion |
| Dead Link Check | 100% Target Link Health | Automated crawl of declared markdown links | Zero 404 Not Found or redirected targets |
| Fast Server TTFB | Sub-50ms Response | Edge SSR & Cache-Control headers | Retrieval completes within sub-second RAG budget |
| Robots Exclusion | RFC-9309 compliance | robots.txt User-agent rules | AI bots granted crawl access to public content |
| CMS Sync | Webhook Revalidation | revalidatePath('/llms.txt') | Updates within seconds of content publish |
How BugViso Audits /llms.txt and Calculates GEO Scores
Because manual checking cannot crawl hundreds of target URLs in real time, auditing your /llms.txt manifest requires automated cloud diagnostics.
+-----------------------------------------------------------------------------------+
| BUGVISO LLMS.TXT SCORING ENGINE |
| |
| [ Web Application Submitted ] ──> [ FastAPI + ARQ Redis Worker Cluster ] |
| │ |
| ▼ |
| [ 4-STAGE LLMS.TXT EVALUATION PIPELINE ] ────────────────────────────────────── |
| ├── 1. Endpoint Resolution QA: Tests HTTP 200 OK and text/plain MIME headers |
| ├── 2. Markdown Syntax Linter: Verifies H1 title, summary, and bullet link syntax|
| ├── 3. Parallel Link Crawler: Crawls 100% of declared URLs to verify 200 OK health|
| └── 4. GEO Citability Engine: Evaluates /llms.txt and AI crawler permissions |
| │ |
| ▼ |
| [ COMPOSITE 0-100 GEO SCORE + ACTIONABLE DEVELOPER REMEDIATION PLAYBOOK ] |
+-----------------------------------------------------------------------------------+When you audit your website on BugViso, the backend crawler executes a complete /llms.txt scoring assessment:
1. Automated Markdown Syntax & Specification Validation
BugViso requests your root /llms.txt and /llms-full.txt files, evaluating formatting against the official specification under Google Search Central Core Web Vitals documentation and W3C Web Content Accessibility Guidelines (WCAG).
2. Deep Broken Link Linter
The engine crawls every URL declared within your /llms.txt manifest in parallel, ensuring that AI crawlers are never directed to 404 Not Found or redirected URLs.
3. Throttled 3G Mobile Performance Simulation
BugViso re-loads pages under CDP Slow 3G (400 ms RTT, 500 Kbps) and Fast 3G network emulation with mobile CPU slowdown, measuring real-world Largest Contentful Paint (LCP) and mobile Interaction to Next Paint (INP).
4. Generative Engine Optimization (GEO) AI Citability Scoring
The platform audits robots.txt AI crawler permissions under RFC 9309 Robots Exclusion Protocol, validates /llms.txt manifests, and calculates your composite 0–100 GEO citability score.
5. Actionable Developer Playbooks & Branded PDFs
Findings are synthesized into a numbered developer remediation playbook in interactive web dashboards and branded ReportLab PDFs. Users receive one full branded PDF report download free every calendar month per device, with on-demand extra reports costing just $4.99.
Frequently Asked Questions About the /llms.txt Audit Template
What is a passing score on the /llms.txt audit template?
A score of 85 points or higher indicates strong AI search readiness, while category leaders achieve 95–100 points.
What is the most common reason for score deductions?
The most frequent issue is broken or redirected target URLs declared within the markdown bullet list, followed by missing /llms-full.txt corpus files.
Does /llms.txt help with Cursor IDE and Claude Projects?
Yes. Providing a valid /llms.txt and /llms-full.txt allows developers using Cursor to ingest your entire documentation portal via @docs in a single command.
How often should we audit our /llms.txt file?
Audit your /llms.txt file monthly or automate validation via CI/CD testing scripts whenever new documentation pages are published.
How can I score my website's /llms.txt file right now?
Run a scan on BugViso to test your /llms.txt endpoint, validate link health, and receive your composite 0–100 GEO citability score.
Conclusion: Elevating Machine Discovery with Certified /llms.txt
Auditing your /llms.txt implementation ensures that your web application delivers seamless, error-free documentation to generative AI search engines and developer tools.
By enforcing standard markdown formatting, publishing an inlined /llms-full.txt corpus, validating target link health, and auditing your manifest with modern cloud diagnostics, engineering teams can guarantee peak machine citability, which is why following this comprehensive llms.txt audit template score guide on BugViso provides the architecture and verification tools needed to build future-proof web applications.
See where your site stands — free.