Keyword gap report API
Topics (free mode) or ranked keywords (licensed mode) a competitor covers that the site does not.
Example
curl -s -X POST https://audit.45-94-66-237.sslip.io/v1/run/seo.keyword_gap \
-H "Authorization: Bearer $SITEAUDIT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"domain":"example.com","competitor":"example.org","limit":10}'
from siteaudit_client import Client
client = Client() # reads SITEAUDIT_API_KEY
result = client.run('seo.keyword_gap', {'domain': 'example.com', 'competitor': 'example.org', 'limit': 10})
print(result['cost_usd'], result['output'])
const res = await fetch("https://audit.45-94-66-237.sslip.io/v1/run/seo.keyword_gap", {
method: "POST",
headers: { Authorization: `Bearer ${process.env.SITEAUDIT_API_KEY}`, "Content-Type": "application/json" },
body: JSON.stringify({"domain": "example.com", "competitor": "example.org", "limit": 10}),
});
const run = await res.json();
console.log(run.cost_usd, run.output);
Sample response
Illustrative and trimmed: a fictional site in the free site mode. Real responses carry more fields; see Output below.
{
"ok": true,
"request_id": "req_b71e0d93c2a54f18",
"endpoint": "seo.keyword_gap",
"status": "succeeded",
"cost_usd": 0.25,
"items": 2,
"latency_ms": 2120,
"source": {
"id": "site",
"name": "First-party crawl",
"kind": "first_party",
"demo": false
},
"output": {
"report_id": "gap_5e2c9a01f4b78d36",
"method": "sitemap_topic_gap",
"summary": {
"total_gaps": 2,
"returned": 2,
"top_opportunity": "invoice templates",
"domain_sitemap_urls": 41,
"competitor_sitemap_urls": 388
},
"section_gaps": [
{
"section": "/templates/",
"competitor_pages": 57
}
],
"gaps": [
{
"keyword": "invoice templates",
"topic": "invoice templates",
"missing_terms": [
"invoice",
"templates"
],
"competitor_url": "https://example.org/templates/invoice",
"volume": null,
"difficulty": null,
"opportunity": null
}
],
"not_covered": [
"Search volume and difficulty (needs a licensed keyword data provider)"
]
}
}
Cost at volume
You pay : $0.25 per report, no plan, nothing when idle. Failed calls are not charged.
Input
| Field | Type | Required | Default |
|---|---|---|---|
domain | string | yes | |
competitor | string | yes | |
location | string | no | "us" |
language | string | no | "en" |
limit | integer | no | 25 |
Output
Inside output of the run envelope:
| Field | Meaning |
|---|---|
report_id | Stable id of the stored report. |
method | sitemap_topic_gap (free mode) or ranked_keywords (licensed mode). |
gaps | Rows the competitor covers and the site does not, best first. |
section_gaps | Free mode: whole site sections only the competitor has. |
sources | Data sources used. |
Method
In the free mode the gap compares both sites' public sitemaps and returns section and topic gaps; the method guide explains what it can and cannot tell you. With a licensed keyword source it returns ranked keywords instead.
Measured performance
No calls measured on this server in the last 30 days yet, so there is no latency or success rate to show. Numbers appear here once real traffic exists; nothing is estimated.
Counted on this server, recomputed on every page view. Not a guarantee or an SLA.
Questions
How much does Keyword gap report cost?
$0.25 per report, drawn from a prepaid USD wallet. The exact amount is on every response as cost_usd, and max_cost_usd lets you refuse anything higher.
Can I check the price before running?
Yes. POST /v1/quote/seo.keyword_gap with the same body validates the input and returns the price and whether your wallet can pay. It is free.
How reliable is it?
The numbers above are measured on this server's own calls over 30 days. There is one data source per mode and no automatic failover; failed calls are refunded.
Can an agent call it?
Yes, as the MCP tool run_endpoint with endpoint: "seo.keyword_gap", or through the CLI and SDKs.
Markdown version for agents: /docs/api/seo.keyword_gap.md
Run the full audit on your site. One call checks every item in the check reference and ranks the fixes.
Open the playground Endpoint details