The Tadaaah API

Why people use it

• One site or many — POST a single URL, or up to 50 per request and poll the whole batch at once.
• Cheap re-audits — repeat domains are served from cache instantly and don't count against your quota.
• Confidence per metric — every score is measured, AI estimate, or estimated, so you know what's externally verifiable.
• CSV export — drop a whole batch straight into a spreadsheet, report, or dashboard.
• Connect any AI agent — a native MCP server plus an OpenAPI spec, so Claude, ChatGPT, Cursor & co. can run audits for you.

  Connect any AI agent (MCP)

Tadaaah speaks the Model Context Protocol — the open standard AI agents use to call external tools. Point Claude, ChatGPT, Cursor or any MCP client at one URL and its agent can run audits, poll results and check your quota by itself:

Authentication is the same API key, sent as an X-API-Key header. Tools exposed: audit_website, get_audit, batch_audit, get_batch, account_usage.

Claude Code (one line)

claude mcp add --transport http tadaaah https://tadaaah.io/mcp \
  --header "X-API-Key: tdh_live_your_key_here"

Claude Desktop, Cursor & other MCP clients (JSON config)

{
  "mcpServers": {
    "tadaaah": {
      "type": "http",
      "url": "https://tadaaah.io/mcp",
      "headers": { "X-API-Key": "tdh_live_your_key_here" }
    }
  }
}

Then just ask your agent: “Audit acme.com and tell me the three most impactful fixes.” Fresh scans take 30–60 s — the agent gets an audit_id and polls get_audit on its own.

Full agent setup guide (ChatGPT, Claude, Cursor…) →

  OpenAPI spec

Building on a framework that ingests OpenAPI instead (custom GPTs, LangChain, n8n, Zapier…)? The whole REST API below is described at:

Authentication

Every request needs your API key in the X-API-Key header (or Authorization: Bearer <key>). PRO subscribers generate a key in one click from the dashboard (100 audits / rolling 30 days). Agencies needing higher volume — contact us for a larger key.

curl https://tadaaah.io/v1/me \
  -H "X-API-Key: tdh_live_your_key_here"

Usage is metered on a rolling 30-day window. Responses include X-RateLimit-Limit and X-RateLimit-Remaining.

Submit a batch

Send one url or many urls (max 50). Optional lang (en/fr) sets the language of the written summary.

curl -X POST https://tadaaah.io/v1/audits \
  -H "X-API-Key: tdh_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://acme.com", "https://beta-client.io"], "lang": "en"}'

Returns 202 with a batch id and one entry per URL. Cached URLs come back done immediately; the rest are pending.

{
  "success": true,
  "batch_id": "9f8c2a...",
  "count": 2,
  "metered": 1,
  "audits": [
    { "url": "https://acme.com",       "audit_id": "a1b2...", "status": "done",    "cached": true  },
    { "url": "https://beta-client.io", "audit_id": "c3d4...", "status": "pending", "cached": false }
  ],
  "quota": { "limit": 5000, "remaining": 4712, "window": "rolling 30 days" }
}

Poll the batch

Returns every audit's current status; completed ones include the full scored result. Poll every few seconds until pending reaches 0.

{
  "success": true,
  "batch_id": "9f8c2a...",
  "total": 2, "completed": 1, "pending": 1,
  "audits": [
    {
      "url": "https://acme.com",
      "status": "done",
      "overall_ai_readiness": 78,
      "scores": {
        "ai_readiness":    { "value": 78, "tier": "ai_estimate", "source": "AI estimate — Tadaaah blended model" },
        "seo":             { "value": 84, "tier": "measured",    "source": "Measured — on-page analysis" },
        "ai_crawlability": { "value": 92, "tier": "measured",    "source": "Measured — raw HTML (no-JS bot view)" },
        "performance":     { "value": 53, "tier": "measured",    "source": "Measured — Google Lighthouse (PageSpeed)" }
      },
      "business": { "industry": "Plumbing", "type": "Local service" },
      "summary": "Acme's site is fast for crawlers but ..."
    }
  ]
}

Single audit

Fetch one audit on its own — same shape as a batch entry.

CSV export

The whole batch as a flat CSV — one row per URL, a column per metric — ready to paste into a client report or a Google Sheet.

curl https://tadaaah.io/v1/batches/9f8c2a....csv \
  -H "X-API-Key: tdh_live_your_key_here" -o audits.csv

Confidence tiers

Mirroring what tadaaah.io shows on every score card, so you can tell clients exactly which numbers are directly verifiable:

Tier	Means	Examples
measured	Read from the live page or a real external tool — externally verifiable.	SEO, AI crawlability, performance (PageSpeed), security (TLS)
ai_estimate	Tadaaah's model judgement.	GEO, AEO, GSO, design, overall AI readiness
estimated	Proxy signals, not a direct measurement.	Local visibility, performance when PageSpeed is unavailable