canonical: https://jentic.com/apis/llmse.ai/llmse

# LLMSE API

The LLMSE API analyzes a website from a single URL, returning a content classification plus separate scores for SEO, E-E-A-T, AEO (Answer Engine Optimization), readability, WCAG accessibility, GARM brand safety, AI disclosure, and privacy compliance. A comprehensive audit runs all of these in one request, fetching the page once and also checking robots.txt, llms.txt, and llms-full.txt for agent readiness. It also matches advertising networks to a page's classification demographics and returns similar sites from the same category and subcategory, and every endpoint works without authentication under per-domain rate limiting.

## For AI agents

Audit any website URL for SEO, E-E-A-T, answer-engine readiness, readability, WCAG accessibility, brand safety, and privacy compliance, classify its content, find similar sites, and match advertisers to it.

## Scope

Does not track keyword rankings, crawl entire sites, or monitor brand mentions in AI answers over time. Use for single-URL website analysis, classification, and advertiser matching only.

## Capabilities

- Run a full website audit with auditUrl that combines classification, SEO, E-E-A-T, AEO, readability, WCAG, GARM, AI disclosure, and privacy checks in one call
- Score a page's answer-engine readiness with analyzeAeo, including checks against robots.txt and llms.txt
- Check a page against WCAG accessibility rules with analyzeWcag and return categorized issues
- Rate brand safety under the GARM framework with analyzeGarm for a URL or a given category and sentiment
- Classify a website's content into a category and subcategory with classifyUrl
- Match advertising networks to a page using its classification demographics or explicit age, gender, and min_cpm filters with selectAdvertiser
- Discover similar websites in the same category and subcategory with findSimilarSites

## Use cases

### Agent-run site audits through Jentic

A coding or marketing agent asks Jentic for the LLMSE operation that audits a website, then calls auditUrl with the page URL. The response bundles classification, SEO, E-E-A-T, AEO, readability, WCAG, GARM, AI disclosure, privacy, advertiser matches, and similar sites, so the agent can turn it into a prioritized fix list. Sub-analyses that fail come back as an error object rather than failing the whole audit.

Example prompt: Search Jentic for 'audit a website for SEO and accessibility', call auditUrl with url 'https://example.com', and list the WCAG and SEO issues it returns

### Answer-engine readiness checks

Before publishing, a team calls analyzeAeo on each key page to score how well it is structured for AI answer engines. The audit variant also fetches robots.txt, llms.txt, and llms-full.txt, so the result shows whether agents and AI crawlers can read the site at all.

Example prompt: Call analyzeAeo with url 'https://example.com/pricing' and report the lowest-scoring metrics

### Brand safety review for ad placements

An ad operations workflow runs analyzeGarm on a publisher URL to score brand safety, then calls selectAdvertiser to rank advertising networks against the page's classification, with optional age, gender, sentiment, and min_cpm filters. Matches are scored on category, age, gender, and sentiment fit, with CPM as tiebreaker.

Example prompt: Call analyzeGarm with url 'https://news.example.com/article', then call selectAdvertiser with the same url and limit 5

### Competitive site discovery

A researcher calls classifyUrl to see how LLMSE categorizes a competitor's site, then findSimilarSites to list other websites in the same category and subcategory, using cached classifications when available. The list seeds a competitive analysis or outreach shortlist.

Example prompt: Call findSimilarSites with url 'https://competitor.example.com' and limit 10, then group the results by subcategory

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | `/api/v1/audit` | Comprehensive audit |
| GET | `/api/v1/classify` | Classify a URL |
| GET | `/api/v1/seo` | Analyze SEO |
| GET | `/api/v1/aeo` | Analyze AEO |
| GET | `/api/v1/wcag` | Analyze WCAG accessibility |
| GET | `/api/v1/garm` | Analyze GARM brand safety |
| GET | `/api/v1/advertisers` | Match advertisers |
| GET | `/api/v1/similar` | Find similar sites |

## Key resources

- **Audit** — One-call combined analysis of a URL across every LLMSE check
- **Analyses** — Individual SEO, E-E-A-T, AEO, readability, WCAG, GARM, AI disclosure, and privacy results
- **Classification** — Category and subcategory for a URL, used for similar sites and advertiser matching

## AI readiness

This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.

- **Score:** 61 / 100
- **Maturity:** AI-Aware
- **Dimensions:**
  - Foundational Compliance: 49 / 100
  - Developer Experience & Jentic Compatibility: 64 / 100
  - AI-Readiness & Agent Experience: 48 / 100
  - Agent Usability: 94 / 100
  - Security: 60 / 100
  - AI Discoverability: 84 / 100
- **View full report:** https://jentic.com/apis/llmse.ai/llmse/scorecard
- **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/
- **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview

### Score it yourself

Every API in the directory is allowlisted, so you can re-score it with no key required.

- **Score your own API:** https://jentic.com/scorecard.md
- **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md

```sh
npx @jentic/api-scorecard-cli score <openapi-url>
```

## Why Jentic

- **Setup:** Wiring LLMSE by hand means handling its optional bearer token, tracking the X-RateLimit headers, and honoring Retry-After on 429 responses yourself. Through Jentic you install once, import LLMSE from the API Directory, and your agent calls it.
- **Permission scoping:** LLMSE takes the target page as a url query parameter, so rules bound which operations your agent may call rather than which sites it analyzes. You choose the operations it may call, so the user-agent submission write is not included unless you add it.
- **Credential handling:** LLMSE endpoints work without a key. If you add a bearer token, it is held encrypted by your own Jentic One instance and injected at execution time, and it never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'audit a website' or 'check brand safety of a URL', and Jentic returns the matching LLMSE operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Moz API** — Link and domain authority metrics for SEO analysis
- **Semrush API** — Keyword, ranking, and site audit data for search marketing
- **Google PageSpeed Insights API** — Lighthouse performance and Core Web Vitals for a URL
- **LLM Pulse API** — Tracks brand mentions and citations in AI-generated answers

## FAQ

### What authentication does the LLMSE API use?

Per its OpenAPI spec, every LLMSE endpoint works without authentication, and an optional HTTP bearer scheme accepts a Descope JWT. Authenticated callers currently receive the same rate limits as anonymous ones. If you do add a token through Jentic, it is stored encrypted by your own Jentic One instance and injected at call time.

### Can I check WCAG accessibility with the LLMSE API?

Yes. Call GET `/api/v1/wcag` with the url query parameter and LLMSE returns a WCAG accessibility result with categorized issues for that page. GET `/api/v1/audit` includes the same WCAG check alongside SEO, E-E-A-T, AEO, and the other analyses.

### What are the rate limits for the LLMSE API?

The OpenAPI spec states per-domain rate limiting of 1 request per domain per 60 seconds, and cached results return without consuming the limit. Responses carry X-RateLimit-Limit, X-RateLimit-Remaining, and X-RateLimit-Reset headers, and a 429 response includes Retry-After.

### How do I run a full website audit with the LLMSE API through Jentic?

Install Jentic One with `curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh`, add the LLMSE API from the Jentic API Directory, and have your agent search for 'audit a website'. Jentic returns auditUrl (GET `/api/v1/audit`) with its schema, and the agent passes the page url. To run it on your own infrastructure, install Jentic One from its GitHub repo.

### Is there an LLMSE MCP server?

You don't need an MCP server to give your agent the LLMSE API. Jentic connects it directly from the API Directory: import it and your agent calls it, with operations discovered on demand instead of loading extra tool definitions into its context.

### Can I limit what my agent is allowed to do with the LLMSE API?

Yes. For an audit-only agent, write a rule that allows auditUrl, analyzeSeo, analyzeAeo, and analyzeWcag. The agent can analyze pages but cannot call submitUserAgent, the one write operation, because it is not in the allowed set, and every call it makes is logged.

### What does the LLMSE comprehensive audit include?

GET `/api/v1/audit` combines classification, SEO, E-E-A-T, AEO with citation and agent readiness, readability, WCAG, GARM, AI disclosure, privacy compliance, advertiser matching, and similar site discovery in one request. It fetches the page once, also reads robots.txt, llms.txt, and llms-full.txt, and returns an error object for any sub-analysis that fails.