Debuly

NHS Agentic Readiness Check

A GitHub Action that fetches your site's agent-readiness score from Not Human Search and can fail a build if it drops below a threshold.

Open source
Visit nothumansearch.ai NHS Agentic Readiness Check (official site)Compare developer platforms

What is NHS Agentic Readiness Check?

NHS Agentic Readiness Check is a GitHub Action for continuous integration. Given a domain, it retrieves the site's agentic readiness score from Not Human Search and can fail the workflow when the score falls under a minimum you set, so a bad deploy that breaks llms.txt or an OpenAPI spec is caught before production. The score covers seven machine-readable signals worth up to one hundred points: llms.txt, an ai-plugin manifest, an OpenAPI spec, a structured API, a verified MCP server, robots.txt AI rules and Schema.org markup. Inputs are the domain, a minimum score with a default of fifty and an optional fresh-crawl request that adds about half a minute. Outputs include the numeric score, a map of detected signals and a link to the report, and a readable summary is posted to the job summary page. The repository's license is not clearly stated.

Use cases

  • Gate deploys on agent readiness
  • Catch broken llms.txt files
  • Track an agent-readiness score in CI
  • Request a fresh crawl before scoring

Who it's for

  • DevOps engineers
  • Web developers
  • Site owners

Why use NHS Agentic Readiness Check

  • Catches regressions in agent-facing files before they reach production.
  • A configurable threshold lets you decide how strict the gate is.
  • The step summary shows per-signal results, so developers see what failed without leaving the workflow.
  • Outputs can feed later steps, such as posting the score to a chat channel.

NHS Agentic Readiness Check features

  • Fetches the Not Human Search score for a domain
  • Fails the build below a minimum score
  • Reports which of seven signals were detected
  • Optional recrawl before scoring
  • Job summary with per-signal breakdown
  • Score, signals and report URL as outputs

How to get started with NHS Agentic Readiness Check

  1. Add the action to a GitHub Actions workflow in your repository.
  2. Set the domain input to your site's host, without the protocol, and optionally a minimum score.
  3. Turn on the recrawl input if you want a fresh crawl first.
  4. Run the workflow and read the job summary, or use the score output in later steps.

NHS Agentic Readiness Check limitations

  • Relies on the Not Human Search service
  • Recrawl adds roughly thirty seconds
  • Measures machine-readable signals, not content quality

NHS Agentic Readiness Check pricing

NHS Agentic Readiness Check is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

NHS Agentic Readiness Check alternatives

The main NHS Agentic Readiness Check alternatives listed on Debuly are Agentprobe, CostGoat, Manifest, Meticulous, MLflow and Serper. The table compares pricing and platforms; order is A–Z.

NHS Agentic Readiness Check compared with its alternatives
ToolPricingStarting pricePlatforms
NHS Agentic Readiness CheckOpen source——
AgentprobeFree—Web, API, GitHub Action
CostGoatFree trial$9/momacOS, Windows, Linux
ManifestOpen source$19/moWeb, Self-hosted, API, Docker
MeticulousPricing unknown—Web, API
MLflowOpen source—Web, Self-hosted, API, CLI
SerperFreemium—API, Web

  1. Agentprobe

    Free scanner that scores a seller website on agentic commerce readiness, showing whether an AI agent can find, quote and buy.

    Developer platformsMarketing and SEOAutomation and agents
    FreeVisit Agentprobe (official site)

  2. CostGoat

    Desktop menubar app that watches AI API credits, agent quotas, cloud bills and SaaS renewals and warns you before spend runs out.

    Developer platformsProductivity
    Free trialVisit CostGoat (official site)

  3. Manifest

    Open source LLM router for agents: one OpenAI- and Anthropic-compatible endpoint with fallbacks, spend limits and automatic request repair.

    Developer platformsAutomation and agentsCoding
    Open sourceVisit Manifest (official site)

  4. Meticulous

    Frontend regression testing that records real user sessions and replays them on every pull request, with no tests to write by hand.

    CodingDeveloper platformsAutomation and agents
    Visit Meticulous (official site)

  5. MLflow

    Open-source AI engineering platform for tracing, evaluating and deploying LLM agents, apps and classic ML models.

    Developer platformsAutomation and agentsCoding
    Open sourceVisit MLflow (official site)

  6. Serper

    Google Search API that returns organic results, knowledge graph and related data as structured JSON in one to two seconds.

    Developer platformsMarketing and SEOResearch and search
    FreemiumVisit Serper (official site)

NHS Agentic Readiness Check FAQ

What is the NHS Agentic Readiness Check?

It is a GitHub Action that pulls your site's agent-readiness score from Not Human Search and can fail CI if the score is below a threshold.

What does the score measure?

It adds up seven machine-readable signals: llms.txt, ai-plugin.json, an OpenAPI spec, a structured API, a verified MCP server, robots.txt AI rules and Schema.org markup.

What is the default minimum score?

The minimum score input defaults to fifty, and you can change it to fit your needs.

Is the action free?

It is a public GitHub repository with no pricing mentioned. Check the repository and Not Human Search for terms.

What are alternatives?

You can run Lighthouse CI for general audits, schema validators in CI, or use the Not Human Search web tool manually. On Debuly, compare Agentprobe, CostGoat and Manifest.

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

SourceTypeAccessed
nothumansearch.ai/api/v1/checkofficial
github.com/unitedideas/nhs-score-check-actionrepository

Image credits

  • Logo: none collected yet, so a monogram is shown.
  • Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.