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

# Mekler AI

The Mekler AI API runs pay-per-call checks for agents, led by x402 conformance scans that grade a paid service origin for protocol and discovery configuration with separate linter and compliance grades. The same API validates CSV deliveries against a JSON column specification, checks citation lists for duplicate and shared source URLs, extracts readable text from static or JavaScript-rendered public pages, inspects an email address or domain for MX, SPF, and DMARC records, and returns annual US-GAAP facts by SEC CIK. Paid calls settle with USDC on Base or Solana through x402 with no account or API key, and the scan status, scan report, and check-list reads are open.

## For AI agents

Scan an x402 service origin for protocol conformance, validate CSV files against a column spec, check citations for shared sources, extract page text, inspect mail-domain DNS, and pull annual SEC facts by CIK, paying per call over x402.

## Scope

Does not crawl whole sites, verify mailbox delivery, repair CSV data, or judge whether a claim is true. Use for x402 conformance scans and bounded per-call data, page, mail-domain, and SEC fact checks only.

## Capabilities

- Scan an x402 service origin for protocol and discovery conformance, then poll the job and read the graded report by report_id
- List both conformance check sets with their grading rules, versions, and evaluation coverage
- Validate a CSV against a JSON column specification, checked on its own first, and get pass or fail with row and column error codes
- Find duplicate citation URLs and shared source links across 2 to 10 supplied citations
- Extract readable text, title, final URL, and source links from a static page, or render a JavaScript page in a headless browser first
- Check an email address or domain for MX, SPF, DMARC, and an optional DKIM selector record
- Retrieve annual US-GAAP financial facts for a company by SEC CIK, with filing links and optional observed revisions

## Use cases

### Agent-run x402 conformance checks through Jentic

A developer shipping a paid x402 endpoint asks their agent to grade it before launch. Through Jentic the agent finds the scan-origin operation, submits the public origin with an idempotency_key, then polls scan status and reads the finished report with separate linter and compliance grades. The status, report, and check-list reads are open; the scan itself is paid per completed scan over x402, and failed or expired work is not charged.

Example prompt: Call POST `/v1/services/conformance/scan-origin` with origin 'https://api.example.com' and a fresh idempotency_key, complete the x402 payment, then poll GET `/v1/services/conformance/retrieve-scan-status` with the returned job_id and the X-Mekler-Scan-Capability header until done, and read GET `/v1/services/conformance/retrieve-scan-report` with the report_id

### CSV delivery validation before import

Data teams receiving vendor CSV files reject bad deliveries before they reach a warehouse. The validate-data operation checks a CSV of up to 2,000 rows and 50 columns against a JSON column specification for missing or extra columns, required cells, string, integer, decimal, boolean, and date types, uniqueness, and allowed values, returning the first 50 failures with row and column codes. The column spec can be checked on its own first.

Example prompt: Call POST `/v1/data/csv/validate-column-spec` with the requirements object, then call POST `/v1/data/csv/validate-data` with the csv text and the returned spec, and report each failing row and column code

### Citation overlap checks for research agents

Research agents counting corroboration need to know when several citations trace back to one source. The source-overlap operation takes a claim label and 2 to 10 citations, with optional article HTML and citation chains, and returns duplicate URLs, shared links, and supplied-source groups with their locations. It does not fetch pages or judge truth, so it measures overlap only.

Example prompt: Call POST `/v1/research/citations/check-source-overlap` with a claim label and four citations, then list any citations that share a source URL

### Mail-domain DNS inspection

Before sending to a new domain or onboarding a sender, an agent checks its published mail configuration. The mail-configuration operation takes kind email or domain and a value, and reports address syntax, MX or null MX with A and AAAA fallback, SPF, exact-domain DMARC, an optional caller-supplied DKIM selector record, and a small known disposable-domain list. It does not verify that a mailbox exists.

Example prompt: Call POST `/v1/email/domains/check-mail-configuration` with kind 'domain', value 'example.com', and selector 'google', and summarize the MX, SPF, DMARC, and DKIM results

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/v1/services/conformance/scan-origin` | Scan an x402 service origin |
| GET | `/v1/services/conformance/retrieve-scan-report` | Read a cached scan report |
| GET | `/v1/services/conformance/list-conformance-checks` | List conformance check sets |
| POST | `/v1/data/csv/validate-data` | Validate a CSV against a column spec |
| POST | `/v1/research/citations/check-source-overlap` | Find shared citation sources |
| POST | `/v1/web/pages/extract-rendered-content` | Render and extract a JavaScript page |
| POST | `/v1/email/domains/check-mail-configuration` | Check mail-domain DNS records |
| POST | `/v1/finance/companies/retrieve-annual-facts` | Retrieve annual SEC financial facts |

## Key resources

- **Conformance scans** — Asynchronous x402 origin scans with job status, cached reports, and the check sets they grade against
- **CSV validation** — Column specifications and CSV deliveries checked against them
- **Web pages** — Readable text and source links extracted from static or JavaScript-rendered public pages
- **Annual financial facts** — US-GAAP annual facts by SEC CIK with filing links and observed revisions

## 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:** 0 / 100
- **Maturity:** Non-Ready
- **Dimensions:**
  - Foundational Compliance: 47 / 100
  - Developer Experience & Jentic Compatibility: 69 / 100
  - AI-Readiness & Agent Experience: 44 / 100
  - Agent Usability: 94 / 100
  - Security: 0 / 100
  - AI Discoverability: 68 / 100
- **View full report:** https://jentic.com/apis/mekler.ai/mekler/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 Mekler AI by hand means handling the x402 flow yourself: send the request unsigned, read the 402 quote, sign one accepts[] offer, and resend the same body with PAYMENT-SIGNATURE, while keeping the scan capability header for recovery. Through Jentic you install once, import Mekler AI from the API Directory, and your agent calls its operations through one gateway.
- **Permission scoping:** Mekler AI carries no resource id in its operation paths, so rules bound which operations your agent may call rather than which origin or file it sends. You choose the operations it may call, so paid work such as conformance scans or headless page rendering is not included unless you add it.
- **Credential handling:** Mekler AI needs no API key: its paid calls settle through x402 payment signatures and its conformance reads are open, so there is no account secret for Jentic to hold. When you add APIs that do use keys, your own Jentic One instance keeps them encrypted and injects them at execution time, never in the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'check an x402 service for conformance' or 'validate a CSV file', and Jentic returns the matching Mekler AI operation with its input schema, including required fields like origin or csv and spec, so the agent calls the right endpoint directly.

## Related APIs

- **402registry** — Audits whether AI crawlers can see a website and sells checks and reports per call over x402
- **Firecrawl API** — Scrapes and crawls websites into clean markdown or structured data
- **AgentFilings API** — SEC filings, earnings, and insider transactions paid per call over x402

## FAQ

### What authentication does the Mekler AI API use?

Per its OpenAPI spec, Mekler AI uses an x402 HTTP scheme, which is a payment rather than an account credential: an unsigned POST returns 402 with a PAYMENT-REQUIRED quote, and you resend the same request with a signed PAYMENT-SIGNATURE header. There is no API key or signup. The three conformance reads (scan status, scan report, and the check list) are open and need no payment, and scan recovery uses the X-Mekler-Scan-Capability header.

### Can I check an x402 service for protocol conformance with Mekler AI?

Yes. POST `/v1/services/conformance/scan-origin` takes a public origin with no path or query, plus optional same-origin paths to probe, and admits an asynchronous scan that returns separate linter and compliance grades. Poll GET `/v1/services/conformance/retrieve-scan-status`, then read the report from GET `/v1/services/conformance/retrieve-scan-report`, which stays cached for 24 hours.

### What are the rate limits for the Mekler AI API?

The OpenAPI spec does not state numeric rate limits, though the paid operations can return 429 Too Many Requests. It does state input bounds, such as CSV validation handling up to 2,000 rows, 50 columns, and 256 KiB. Check the Mekler documentation at https://mekler.ai for current limits.

### Is there a Mekler AI MCP server?

You don't need an MCP server to give your agent Mekler AI. Jentic connects it directly from the API Directory: import it and your agent calls the scan, validation, and extraction operations it needs on demand, without loading another server's tool definitions into its context.

### How do I validate a CSV file through Jentic with Mekler AI?

Search Jentic for 'validate a CSV against a column spec', load the POST `/v1/data/csv/validate-data` operation, and execute it with the csv text and a spec object describing the columns. The response is pass or fail with row and column error codes for the first 50 failures. To run it on your own infrastructure, install Jentic One from its GitHub repo.

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

Yes. Write a rule that allows only the CSV validation and column-spec check operations, and your agent can vet data deliveries but cannot start paid conformance scans, render pages, or pull SEC facts. Each call it makes is logged, so you can review which checks it ran and paid for.