{
  "markdown": "# Website Auditor MCP\n\nAn [MCP](https://modelcontextprotocol.io) server for **[website-auditor.io](https://website-auditor.io)** —\nAI-visibility (GEO) and site-audit tools you can call from any MCP client\n(Claude Desktop, Claude Code, Cursor, Codex, and other agents).\n\nAsk an agent *\"does ChatGPT recommend my business?\"*, *\"what's my AI-visibility\nscore?\"*, *\"audit example.com\"*, or *\"how do I stack up against my competitors?\"*\nand it answers with real data — an overall AI-visibility score (0–100), a\nper-engine breakdown across **ChatGPT, Perplexity, Claude and Gemini**, a full\nsite audit (SEO, security, performance), competitor comparisons, and ongoing\nmonitoring.\n\nThe server is a thin, authenticated wrapper over the Website Auditor API — the\naudit engine, AI-visibility scoring and monitoring live in that service; this\nserver just makes them available to agents.\n\n---\n\n## Tools\n\n| Tool | Tier | What it does |\n|---|---|---|\n| `get_sample_audit` | **Free — no key** | A complete sample report for `example.com` in the exact shape a real audit returns. Needs no API key, no subscription and no network. Try this first to see what you'd be buying. |\n| `get_ai_visibility` | **Pro** | Current AI-visibility score (0–100) + per-engine breakdown (ChatGPT, Perplexity, Claude, Gemini) + the top competitor appearing in your place. Pro subscribers also get `trend`: 7- and 30-day score movement from stored snapshot history. |\n| `run_audit` | **Pro**, rate-limited | Full one-time audit → category scores (AI visibility, SEO, security, performance) + top issues + a shareable report URL. |\n| `get_changes` | **Pro** | What changed since the last check — score movement, engines gained/lost, competitor moves, new/resolved issues. Requires the domain to be tracked. |\n| `compare_competitors` | **Pro** | Head-to-head AI-visibility ranking against named competitor domains + where each appears that you don't. Quota-aware: caps the audit fan-out to your remaining daily quota, reuses recent cached audits, and reports any competitors it had to skip rather than dropping them silently. |\n| `track_site` | **Pro** | Start (or stop) weekly monitoring of a site's AI visibility. Establishes the history `get_changes` reads from. |\n| `untrack_site` | **Pro** | Stop monitoring a site and free up a monitoring slot. Idempotent. |\n| `list_tracked_sites` | **Pro** | List the sites you're monitoring, with cadence, active state, and slots used/remaining. |\n| `get_monitoring_status` | **Pro** | A glanceable dashboard across all tracked sites — latest score, when each was last checked and next runs, and the most recent change. |\n| `check_upgrade_status` | Any valid key | Your own subscription standing — tier, status, period end, and what upgrading unlocks (starting Pro requires a payment method and accepting the Terms). Consumes no audit quota. |\n\n## Prompts\n\nClients that support MCP prompts (Claude Desktop, claude.ai) render these as\nsomething you can pick from a menu, so you don't have to phrase the request\nyourself or know which tool to ask for. In clients that don't surface prompts\n(Codex, currently), nothing is lost but the menu — ask in words and the same\ntools run.\n\n| Prompt | Needs a key? | What it does |\n|---|---|---|\n| **See a sample report** | **No** | Walks through a complete report for `example.com`. No arguments, no setup — one click from any install. |\n| **Check my AI visibility** | Pro | Runs `get_ai_visibility` for a domain you name, then explains the score, which assistants name the business, and who is named instead. |\n| **Run a full site audit** | Pro | Runs `run_audit` for a domain and summarises it by category, with the three fixes that matter most. |\n| **Compare me to a competitor** | Pro | Runs `compare_competitors` for your domain against a named rival, and explains where they get named and you don't. |\n\nEach Pro prompt falls back to `get_sample_audit` when no API key is configured,\nso you always get output rather than an error.\n\n### Naming the business (optional)\n\n`get_ai_visibility` and `run_audit` both accept two optional arguments that\ndecide *what question* the AI-visibility check actually asks.\n`compare_competitors` accepts `business_location` only — see below for why it\ntakes no name:\n\n| Argument | Omitted | Supplied |\n|---|---|---|\n| `business_name` | Detected from the site, and flagged with `name_warning` when it could not be verified | Taken as fact and recorded as caller-supplied — which suppresses the warning |\n| `business_location` | Detected from the site; if nothing is found the questions widen to the country, or drop the place entirely | Scopes the questions to that place |\n\n**Supply a name only when you actually know it.** A supplied name overrides\ndetection and is treated as confirmed, so a guess is scored exactly as if a\nhuman had verified it — and silences the warning that would have told you\notherwise. Leaving it out is the safer default: detection is transparent about\nits own uncertainty.\n\nThe same applies to location, in the other direction. Omitting it is correct\nfor a national or global business and wrong for a local one, since a local\nbusiness measured without a place is measured against the wrong queries.\n\n**On `compare_competitors`:** it takes `business_location` and applies it to\nyour site *and every competitor*, because a comparison asks about one market —\nscoring you locally while scoring your rivals globally would rank the answers\nto two different questions. It deliberately takes no `business_name`: a name\nidentifies one business, and forwarding yours to your competitors would score\nevery one of them as you. The result names the market it used, so a scoped\ncomparison is never mistaken for a global one.\n\n```text\n\"Check AI visibility for hawaiibackroad.com,\n the business is Big Island Backroad Adventures in Hilo, HI\"\n```\n\n---\n\n## Install & configure\n\nThe server runs directly via `npx` — no clone or build required.\n\n**Try it before you buy it.** Install with no API key at all and ask your agent\nfor a *sample audit* — `get_sample_audit` returns a full report for `example.com`\nin the exact format a real run produces, so you can check the shape fits your\nneeds first.\n\n**Pricing.** Auditing real domains needs a Website Auditor subscription at\n**$10/month** — eligible new customers get a **7-day free trial** (payment\nmethod required to start; no charge until the trial ends; customers who used\na trial in the last 12 months are billed immediately). Sign up and create an\nAPI key at\n**[api.website-auditor.io/admin_portal](https://api.website-auditor.io/admin_portal/?source=mcp)**,\nthen set it as `WA_API_KEY` below. There is no free API tier — a key only\nfunctions with an active subscription.\n\n**Claude Desktop** (`claude_desktop_config.json`), **Cursor**\n(`~/.cursor/mcp.json`), and most other clients use the same `mcpServers` shape:\n\n```jsonc\n{\n  \"mcpServers\": {\n    \"website-auditor\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"website-auditor-mcp\"],\n      \"env\": {\n        \"WA_API_KEY\": \"wa_your_key_here\"\n      }\n    }\n  }\n}\n```\n\n**Claude Code** — add it from the CLI:\n\n```bash\nclaude mcp add website-auditor -e WA_API_KEY=wa_your_key_here -- npx -y website-auditor-mcp\n```\n\n**Codex** — the CLI, IDE extension and ChatGPT desktop app all read the same\n`~/.codex/config.toml`, so one of these covers all three:\n\n```bash\ncodex mcp add website-auditor --env WA_API_KEY=wa_your_key_here -- npx -y website-auditor-mcp\n```\n\n```toml\n[mcp_servers.website-auditor]\ncommand = \"npx\"\nargs = [\"-y\", \"website-auditor-mcp\"]\n\n[mcp_servers.website-auditor.env]\nWA_API_KEY = \"wa_your_key_here\"\n```\n\nCodex doesn't render MCP prompts, so the [Prompts](#prompts) above won't appear\nas menu entries there — ask in words instead (*\"show me a sample audit\"*).\n\nOr install the packaged **Codex plugin**, which bundles the same server and\nrestores the prompts as skills (which Codex does render):\n\n```bash\ncodex plugin marketplace add SpikeyCoder/website-auditor-mcp\ncodex plugin add website-auditor@spikeycoder\n```\n\n**Cursor** — one-click install, or put the same `mcpServers` JSON as above in\n`~/.cursor/mcp.json`:\n\n[![Install in Cursor](https://cursor.com/deeplink/mcp-install-dark.svg)](https://cursor.com/install-mcp?name=website-auditor&config=eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIndlYnNpdGUtYXVkaXRvci1tY3AiXSwiZW52Ijp7IldBX0FQSV9LRVkiOiIifX0%3D)\n\nThe one-click config arrives with an empty `WA_API_KEY` — the sample report\nworks as-is; fill the key in under **Settings → MCP** to audit real domains.\nThere is also a packaged **Cursor plugin** (this server plus the prompts as\nskills) in [`cursor-plugin/`](cursor-plugin/); its Cursor Marketplace listing\nis pending — status in [docs/CURSOR-PLUGIN.md](docs/CURSOR-PLUGIN.md).\n\nRestart the client and the tools appear.\n\n### Getting an API key\n\n`WA_API_KEY` is a per-user key (it starts with `wa_`) minted from a Website\nAuditor account at\n**[api.website-auditor.io/admin_portal](https://api.website-auditor.io/admin_portal/?source=mcp)**\n— the admin portal, where you subscribe and manage keys.\n\nMinting a key requires an active subscription ($10/month; eligible new\ncustomers get a 7-day free trial — payment method required, no charge until\nthe trial ends): there is no free API tier, so every tool except\n`get_sample_audit` and `check_upgrade_status` needs one. `get_sample_audit`\nneeds no key at all.\n\nTreat the key like a password — set it only in your MCP client's `env` and never\ncommit it.\n\n**Restart after setting or changing the key.** `WA_API_KEY` is read once, when\nthe server starts, so a key added while the client is running is invisible to\nit — in Claude Desktop, quit and reopen the app. Without the restart the tools\nkeep returning the same `AUTH_REQUIRED` you just acted on, which looks\nidentical to the key not working.\n\n### Configuration (environment variables)\n\n| Var | Default | Purpose |\n|---|---|---|\n| `WA_API_KEY` | _(required)_ | Per-user API key (starts with `wa_`). |\n| `WA_API_BASE_URL` | `https://api.website-auditor.io` | The Website Auditor API this server wraps. |\n| `WA_SITE_URL` | `https://website-auditor.io` | Used to build shareable report links. |\n| `WA_UPGRADE_URL` | `https://api.website-auditor.io/admin_portal/` | Where auth and subscription errors point you. `?source=mcp` is appended so a signup that started here is attributable; set your own `source` to override. Not surfaced on quota errors — the daily cap is not an upsell. |\n| `WA_REQUEST_TIMEOUT_MS` | `120000` | Timeout for API calls. |\n| `WA_AUDIT_CACHE_TTL_MS` | `86400000` | Reuse a domain's audit within this window instead of spending quota (used by `compare_competitors`). Defaults to 24h. |\n| `WA_SUBSCRIPTION_CACHE_TTL_MS` | `60000` | How long a resolved Pro/free tier is cached per key before re-checking the subscription. |\n| `WA_METRICS_DISABLED` | _(unset → metrics on)_ | Set to `1`/`true` to disable anonymous usage telemetry. |\n\nOnly `WA_API_KEY` is normally needed; the rest have sensible defaults. See\n[`.env.example`](.env.example) for the full list.\n\n---\n\n## Auth & tiers\n\nYour key is validated on every call. The Pro/free tier is resolved live from the\nAPI and cached briefly, so upgrades and downgrades take effect within about a\nminute:\n\n- **No key** → `get_sample_audit` still works (that's the point of it); every\n  other tool returns `AUTH_REQUIRED` with the price and a sign-up link.\n- **Revoked or unrecognized key** → `INVALID_KEY`, carrying the API's own\n  remediation (\"generate a new key\"). Distinct from `PRO_REQUIRED`: the fix is a\n  new key, not a purchase.\n- **No active subscription** (valid key, lapsed/canceled/never subscribed) →\n  `PRO_REQUIRED` with the price and an upgrade link — there is no free API tier;\n  `check_upgrade_status` still answers so the caller can learn why.\n- **Subscribed** (status `active` or a trial in progress) → all tools.\n\nErrors are normalized to stable codes agents can branch on — e.g.\n`AUTH_REQUIRED`, `INVALID_KEY`, `PRO_REQUIRED`, `OVER_QUOTA`,\n`UNREACHABLE_DOMAIN`, `INVALID_INPUT`, `TIMEOUT`. A domain that can't be reached\nreturns `UNREACHABLE_DOMAIN` — never a fabricated score.\n\n---\n\n## Develop\n\n```bash\nnpm install\nnpm run build      # compile TypeScript → dist/\nnpm start          # serve over stdio\nnpm run dev        # run from source without building\nnpm test           # vitest\nnpm run typecheck\n```\n\nThe suite is hermetic — the API client, subscription provider and meter are\ninjected, and HTTP is mocked at the `fetch` boundary, so no network is touched.\n\n---\n\n## Privacy Policy\n\n**Anonymous install id.** When telemetry is enabled, the server generates a\nrandom UUID on first run and stores it at\n`~/.config/website-auditor-mcp/install-id` (or `$XDG_CONFIG_HOME`), sending it\nwith each event. It exists solely to tell one install restarting many times\napart from many separate installs — without it, install counts are just restart\ncounts. It is randomly generated, never derived from your machine, username or\nnetwork, and is not a fingerprint. Setting `WA_METRICS_DISABLED` stops\ntelemetry entirely: no id is generated and nothing is written to disk.\n\nThis connector talks to a single external service: the **Website Auditor API**\nat **[website-auditor.io](https://website-auditor.io)**. When you invoke a tool\nit sends only two things to that API:\n\n- the **target domain** you asked to audit or monitor, and\n- your **API key** (`WA_API_KEY`), used to authenticate the request and resolve\n  your plan tier.\n\nThat's the full extent of what leaves your machine. The connector does **not**\ncollect, store, or transmit your files, prompts, conversation content, or any\nother personal data, and it does not send data to any third party beyond the\nWebsite Auditor API. Your API key is held only in your MCP client's\nconfiguration (in Claude Desktop it is stored in the OS keychain and injected as\nan environment variable); it is never written to the bundle or logged.\n\nAnonymous, aggregate usage telemetry (which tool ran, success/failure, latency —\nno domains, no keys, no personal data) may be emitted to improve the service, and\ncan be disabled entirely by setting `WA_METRICS_DISABLED=1`. Installed as a\ndesktop extension, the same opt-out is a **\"Disable anonymous usage telemetry\"**\ncheckbox in the extension's settings.\n\nFull privacy policy: **https://website-auditor.io/privacy**\n\n---\n\n## License\n\n[MIT](LICENSE) — © 2026 Kevin Armstrong / SpikeyCoder.\n\nThis covers the MCP server in this repo: the client that talks to the Website\nAuditor API. The audit engine and the API behind it are separate products, not\ncovered here, and running real audits still needs a Website Auditor account.\n\nLearn more at **[website-auditor.io](https://website-auditor.io)**.\n",
  "bytes": 14718,
  "sha": "1f41d040477c6e1174e004a5673a0ce04928aa1049b8f6fbe464d4d2d1deefbc",
  "repo_slug": "spikeycoder/website-auditor-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_spikeycoder_website_auditor_mc_7b590554/readme"
}