scanned Jun 30, 2026

Paul Graham

paulgraham.com

The site appears to be a personal website or blog of Paul Graham, providing information and resources related to startups and entrepreneurship.

0/100

Tier 1 · Agent-Unreadable

Content answers0/100

Protocol plumbing20/1003 of 15 checks pass

Re-audit632h ago· 1 credit

Scored by asking 10 questions a buyer of a education product asks, then grading this site’s own pages: answered, hedged (partial or vague), or silent (no page answers it). How scoring works

10 buyer questions · graded against the site’s own pages

0 of 10 come back answered. 10 meet silence.

0 answerable0 half-answers10 silent

Goes silent · content · high importance

What specific subjects—like fundraising, hiring, or idea generation—do the essays cover?

Goes silent · navigation · high importance

Can I search the full text of all essays by keyword, or do I have to browse the archive manually?

This report is public. Own paulgraham.com? Claiming is free: crawl every page, re-audit as you fix, and track your score over time.

The fix queue

100 points sit between paulgraham.com and 100: 10 open questions and 12 missing protocol checks, ordered by estimated payoff.

Point estimates are per fix under scoring v2. They are not additive to a promised total.

01content · importance highGoes silent+10 content pts est.

What specific subjects—like fundraising, hiring, or idea generation—do the essays cover?

What the pages say

The retrieved chunks do not contain information about what specific subjects the essays cover. They only display a site header with links to two recent essays (“How to Earn a Billion Dollars” and “The Brand Age”) and a reference to Y Combinator.

The fix

Add an essays-by-topic index or categories page that lists covered subjects like fundraising, hiring, and idea generation.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

02navigation · importance highGoes silent+10 content pts est.

Can I search the full text of all essays by keyword, or do I have to browse the archive manually?

What the pages say

No page on the site addresses this.

The fix

Add a help or navigation page that explains how users can find essays, including whether full-text keyword search is available or if browsing the archive manually is required.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

04author · importance mediumGoes silent+10 content pts est.

What is Paul Graham's current role and past experience with Y Combinator and startups?

What the pages say

The retrieved chunks do not contain information about Paul Graham's current role or past experience with Y Combinator and startups.

The fix

Add an About or bio page that details Paul Graham's current role and his history with Y Combinator and startups.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

05support · importance mediumGoes silent+10 content pts est.

Is there a public email address or preferred channel for readers to send feedback or startup questions?

What the pages say

The retrieved chunks do not contain a public email address or preferred channel for reader feedback or startup questions.

The fix

Add a contact page, footer link, or 'About' section with a public email address or preferred channel (e.g., Twitter, forum) for reader feedback and startup questions.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

06recency · importance mediumGoes silent+10 content pts est.

How often are new essays typically published, and is there a predictable schedule or announcement pattern?

What the pages say

The retrieved chunks do not indicate how often new essays are published or whether there is a predictable schedule or announcement pattern.

The fix

Add an FAQ or about page that explains the essay publishing cadence and how new essays are announced.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

07getting-started · importance mediumGoes silent+10 content pts est.

Is there a curated list of the most important essays for someone new to startups?

What the pages say

No page on the site addresses this.

The fix

Create a dedicated 'Getting Started' or 'Best Essays for Founders' page that curates the most important essays for newcomers.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Page missing

08operations · importance lowGoes silent+10 content pts est.

Are older essays updated with corrections or new context, or are they preserved exactly as originally written?

What the pages say

The retrieved chunks do not address whether older essays are updated with corrections or preserved as originally written.

Source: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

09syndication · importance lowGoes silent+10 content pts est.

Does the RSS feed provide the full essay text, or only a title and link summary?

What the pages say

No page on the site addresses this.

The fix

Add a dedicated page or footer note describing the RSS feed and whether it includes full essay text or only summaries.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Absent

10archive · importance lowGoes silent+10 content pts est.

Are the pre-Y Combinator essays on Lisp and programming still accessible in the same archive as the startup content?

What the pages say

No page on the site addresses this.

The fix

Add a dedicated archive or FAQ page that explains how pre-Y Combinator Lisp essays are organized and whether they are stored alongside startup content.

Where we looked: /llms.txt, paulgraham.com

confidence high · grounding synthesized · weight 0.00 · Page missing

Protocol plumbing · 20/1003 of 15 checks pass · each fix +7 protocol pts est.

The other half of the score: 15 checks for the files and headers agents look for. The 12 below are installs, not judgment calls, and most are an afternoon. Expand any for the snippet and the standard it follows. They sit after the queue because none of them changes what your pages say.

sitemap.xmlDiscoverability+7 pts est.

Standardsitemaps.orgIndustry standard

llms.txtDiscoverability+7 pts est.

Sitedex generates this file from your crawl. Grab it in Files from this audit below.

Standardllmstxt.orgCommunity spec

Content signalAccess+7 pts est.

Install snippet

User-agent: *
Content-Signal: search=yes, ai-input=yes, ai-train=no
Allow: /

StandardCloudflare proposalVendor proposal

Markdown negotiationRendering+7 pts est.

StandardRFC 9110 + 7763IETF RFC

Server-rendered contentRendering+7 pts est.

StandardSitedex metricSitedex metric

MCP cardInteraction+7 pts est.

Sitedex generates this file from your crawl. Grab it in Files from this audit below.

StandardModel Context ProtocolCommunity spec

OpenAPI specInteraction+7 pts est.

StandardOpenAPI SpecIndustry standard

WebMCP widgetInteraction+7 pts est.

Sitedex generates this file from your crawl. Grab it in Files from this audit below.

StandardW3C WebMCP draftW3C / WHATWG

Canonical URLsHygiene+7 pts est.

Install snippet

<link rel="canonical" href="https://paulgraham.com/" />

StandardRFC 6596IETF RFC

Meta descriptionsHygiene+7 pts est.

Install snippet

<meta name="description" content="paulgraham.com: [outcome you deliver] for [who you help]. One sentence, 50-160 characters." />

StandardHTML Living StandardW3C / WHATWG

HTML lang attributeHygiene+7 pts est.

Install snippet

<html lang="en">

StandardHTML LS + WCAG 2.1W3C / WHATWG

Organization schemaIdentity+7 pts est.

Sitedex generates this file from your crawl. Grab it in Files from this audit below.

StandardSchema.org + JSON-LDIndustry standard

Already passing 3 of 15: robots.txt, AI crawler access, Clean crawl.

Ask this site’s index

Sitedex already serves paulgraham.com as an MCP endpoint. Ask paulgraham.com anything an AI agent might ask, and see what its index returns. (To score your own site, use the form below.)

Snippets & configs

For developers and the engineer-on-call: copy these into your tools or your site.

Files from this audit

Built from this crawl. Download or copy each, then install it at the path noted.

llms.txt

Built from this crawl. Install at /llms.txt so agents start here.

organization.json

Organization JSON-LD, pre-filled from this crawl. Wrap in a ld+json script.

server-card.json

MCP server card built from this crawl. Host at /.well-known/mcp/server-card.json.

webmcp.json

WebMCP discovery manifest built from this crawl. Host at /.well-known/webmcp.json.

MCP endpoint

https://mcp.sitedex.dev/s/paulgraham-com/mcp

The URL anyone's agent points at. Read-only; safe to share.

Claude Code

claude mcp add paulgraham --transport http https://mcp.sitedex.dev/s/paulgraham-com/mcp

One command, then the agent has it.

Cursor / Continue

{
  "mcpServers": {
    "paulgraham": {
      "url": "https://mcp.sitedex.dev/s/paulgraham-com/mcp"
    }
  }
}

Drop into mcp.json.

WebMCP: two parts

WebMCP-capable browsers run the widget at runtime. Crawlers without JS rendering need the discovery manifest to find your tool surface. Install both.

1 · Widget script

<script async src="https://sitedex.dev/widget.js"></script>

Drop in <head>. WebMCP-capable browsers (Chrome 146+ Origin Trial) call navigator.modelContext.provideContext() via this script.

2 · Discovery manifest

{
  "$schema": "https://wellknownmcp.org/schemas/webmcp.json",
  "name": "paulgraham.com",
  "tools": [
    { "name": "search", "description": "Search paulgraham.com's indexed content." },
    { "name": "get_page", "description": "Fetch a page from paulgraham.com as markdown." }
  ]
}

Host alongside the script at /.well-known/webmcp.json. Crawlers that don't render JS rely on this.

Your turn

See which of these questions your site goes silent on.

Free, about 5 minutes. We crawl your site, test it against the buyer questions your category asks, and name what’s vague, contradictory, or missing, plus the files AI agents look for.

ComingEmbeddable grade badgeScore history and deltasOpt-in public board