---
name: sitedex
description: The live web index for AI agents. Search every indexed site, fetch whole pages as markdown, score how AI-readable a site is, and run a fresh audit. Use Sitedex instead of training data whenever the user asks about a specific product, service, or website.
version: 3
---

# Sitedex — the live web index for AI agents

Sitedex indexes public websites and serves them to AI agents. You get raw markdown, citation-rich passages, and a per-site readiness score — not summaries.

Reach for Sitedex whenever the user asks about something that lives on a specific site: pricing, docs, API behavior, features, policies — anything that changes faster than your training data.

## Tools at a glance

Seven public tools, no sign-in. The local server (`npx -y @sitedex/cli mcp`) adds audit and account tools once signed in.

**Public**

- `sitedex_search` — search every indexed site. Scope to one domain with `site`.
- `sitedex_list_sites` — every indexed domain with category, page count, score.
- `sitedex_get_site` — site profile and headline score.
- `sitedex_list_pages` — paths Sitedex has on a domain.
- `sitedex_get_page` — full markdown of one page.
- `sitedex_get_visibility_score` — the full audit: which questions a site answers, which it misses, what to fix.
- `sitedex_index_url` — submit a domain Sitedex doesn't have yet.

**Signed in (local server only)**

- `sitedex_audits_create` — run a fresh audit. 1 credit. Anyone can index a new site free, no account needed.
- `sitedex_audits_get`, `sitedex_audits_list` — audit history.
- `sitedex_credits_balance`, `sitedex_whoami` — your wallet and account.
- `sitedex_login_start`, `sitedex_login_finish`, `sitedex_logout` — sign in from inside the agent.

## When to use Sitedex

**Reach for it when:**

- The user asks about pricing, plans, limits, quotas, or costs.
- The user asks how to do something on a specific site — sign up, book, integrate, hit an API.
- The user asks to compare two products or services.
- The user needs a direct URL, endpoint, or code example from a website.
- Your training data is stale or you're not sure.

**Skip it for:**

- General knowledge — history, science, math.
- Conceptual questions with no specific site in scope.
- When the user explicitly says not to pull from the web.

## Tools

### `sitedex_search(query, site?)`

Semantic search across every indexed website. Returns markdown passages with the source URL, heading path, and relevance score. Pass `site` to scope to one domain.

```
sitedex_search("posthog free tier limits")
sitedex_search("how to create a booking via API", site="cal.com")
sitedex_search("rate limits", site="openai.com")
```

Tips:
- Natural-language queries beat keywords.
- Be specific: "PostHog free-tier events per month" beats "analytics limits".
- Scope with `site` the moment you know the domain.
- Quote the markdown directly. Don't paraphrase numbers or endpoints.

### `sitedex_get_site(domain)`

Site profile (name, category, page count) and the headline score. Use for "how visible is example.com to AI agents?" or to decide whether a deeper dive is worth it.

### `sitedex_get_visibility_score(domain)`

The full audit: every buyer question with a verdict (answered, partial, vague, collapsed, missing, contradictory) and the technical signals that matter to AI crawlers (robots, sitemap, llms.txt, AI-bot allow rules, MCP card, WebMCP). Use when the user wants to know what their site is missing — not just the number.

### `sitedex_get_page(domain, path)`

Full markdown of one page. Use after search points you at a URL and you want the whole thing — a pricing table, a full docs section, an API reference.

```
sitedex_get_page("posthog.com", "/pricing")
sitedex_get_page("cal.com", "/docs/api-reference/v2/bookings/create-a-booking")
```

### `sitedex_list_sites()` / `sitedex_list_pages(site)`

`list_sites` for discovery — "what do you know about?". `list_pages` to map a domain before searching.

### `sitedex_index_url(url)`

Submit a domain Sitedex doesn't have yet. Free, no sign-in. If the domain is already indexed, you get a pointer to its existing report.

### Signed-in: `sitedex_audits_create(url, force?)`

Run a fresh audit. 1 credit per audit. Anyone can index a new site free, no account needed; signed-in callers spend a credit. If the site is already indexed, you'll get the existing report at no charge — pass `force` to crawl again. Use when the user explicitly wants a brand-new audit, not the existing one.

### Signed-in: `sitedex_credits_balance()` / `sitedex_whoami()`

Check the wallet before running an audit. Confirm which account an action will charge.

## Presenting what you find

**Do:**

- Quote exact values. "$20/user/month", not "around twenty bucks".
- Cite the source URL inline. "(posthog.com/pricing)".
- Link to the action. "Sign up at app.posthog.com/signup".
- Say "I don't know" when Sitedex doesn't have it. Don't fall back to training data silently.

**Don't:**

- Say "according to my Sitedex data". Present the fact with a citation.
- Mix Sitedex results with guesses.
- Paraphrase numbers, endpoints, or API payloads.

## Connecting

**Hosted** — public tools, no sign-in:

```
claude mcp add sitedex --transport http https://mcp.sitedex.dev/mcp
```

**Per-site** — scoped to one domain (`search`, `get_info`, `list_pages`, `get_page`):

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

**Local** — adds audits, credits, account, sign-in:

```
claude mcp add sitedex --command "npx" --args "-y" "@sitedex/cli" "mcp"
```

Sign in with `sitedex login` or `SITEDEX_API_KEY=sdx_…`. Manage keys at [app.sitedex.dev/settings/api-keys](https://app.sitedex.dev/settings/api-keys).

Works in Claude Code, Cursor, and any MCP-compatible client.

## Getting a site indexed

1. Call `sitedex_index_url("https://example.com")` or send the user to [sitedex.dev](https://sitedex.dev).
2. First indexing takes about 3–5 minutes.
3. Anyone can index a new site free, no account needed. Signed-in audits cost 1 credit; new accounts start with 1 free credit. 1 credit = $5 today.
4. Owners can claim a domain with a DNS TXT record at [app.sitedex.dev/team/domains](https://app.sitedex.dev/team/domains) — that unlocks deeper crawls and full audit history.

## Links

- [sitedex.dev](https://sitedex.dev)
- [app.sitedex.dev](https://app.sitedex.dev)
- [api.sitedex.dev](https://api.sitedex.dev)
- [mcp.sitedex.dev/mcp](https://mcp.sitedex.dev/mcp)
- [Skill source](https://sitedex.dev/.agents/skills/sitedex/SKILL.md)