Title: call for testing – WordPress AI

---

#  Tag Archives: call for testing

 [  ](https://profiles.wordpress.org/laurisaarni/) [Lauri Saarni](https://profiles.wordpress.org/laurisaarni/)
6:50 am _on_ 25 March 2026     
Tags: [AI Experiments ( 16 )](https://make.wordpress.org/ai/tag/ai-experiments/),
call for testing, [core-ai ( 78 )](https://make.wordpress.org/ai/tag/core-ai/), 
[needs testing ( 2 )](https://make.wordpress.org/ai/tag/needs-testing/)   

# 󠀁[Call for Testing: Community AI Connector Plugins](https://make.wordpress.org/ai/2026/03/25/call-for-testing-community-ai-connector-plugins/)󠁿

The WordPress AI Team is excited to invite the community to test a growing collection
of **community-built AI connector plugins**. These plugins extend WordPress by connecting
it to additional AI services through the [PHP AI Client](https://github.com/WordPress/php-ai-client)—
the provider-agnosticProvider-Agnostic Software design that works with multiple 
service providers without being tied to one. Recommended for WordPress AI integrations.
SDK that is part of the upcoming WordPress 7.0 release (read more about it [here](https://make.wordpress.org/core/2026/03/24/introducing-the-ai-client-in-wordpress-7-0/)).

WordPress 7.0 will ship with the PHPPHP PHP (recursive acronym for PHP: Hypertext
Preprocessor) is a widely-used open source general-purpose scripting language that
is especially suited for web development and can be embedded into HTML. [https://www.php.net/manual/en/index.php](https://www.php.net/manual/en/index.php)
AI Client baked into coreCore Core is the set of software required to run WordPress.
The Core Development Team builds WordPress., and connectors for Anthropic, Google,
and OpenAI appear as default options in the connector screen — ready to install 
with a single click. But the real power of the PHP AI Client’s open architecture
is that **anyone can build a connector for any AI service**. The community has already
stepped up — and these plugins are ready for testing.

If you’ve been curious about WordPress’s evolving AI capabilities, this is a great
opportunity to try them out firsthand and help shape the experience.

## What are AI connector plugins?

AI connector plugins act as bridges between WordPress and external AI services. 
They implement the PHP AI Client’s providerProvider An AI service offering models
for generation, embeddings, or other capabilities (e.g., Anthropic, Google, OpenAI).
interface, allowing any WordPress pluginPlugin A plugin is a piece of software containing
a group of functions that can be added to a WordPress website. They can extend functionality
or add new features to your WordPress websites. WordPress plugins are written in
the PHP programming language and integrate seamlessly with WordPress. These can 
be free in the WordPress.org Plugin Directory [https://wordpress.org/plugins/](https://wordpress.org/plugins/)
or can be cost-based plugin from a third-party. or theme to use AI capabilities —
text generation, image generation, function calling, and more — through a unified,
provider-agnostic APIAPI An API or Application Programming Interface is a software
intermediary that allows programs to interact with each other and share data in 
limited, clearly defined ways..

Connectors for **Anthropic**, **Google**, and **OpenAI** are featured prominently
in the WordPress 7.0 connector screen and can be installed directly from there. 
The community connector plugins featured in this article expand the ecosystem further,
giving users and developers more choice in which AI services power their WordPress
sites.

Each connector plugin handles the specifics of communicating with its respective
AI service, so developers and users can switch between providers without changing
their code or workflow.

![](https://make.wordpress.org/ai/files/2026/03/connector-screen-mistral-1024x573.
jpg)

Screenshot of the connector screen having a community plugin connected.

## Community connector plugins to test

### AI Provider for Grok (xAI)

[Grok](https://x.ai/) is xAI’s conversational AI, known for its real-time knowledge
and distinctive personality.

 * **Author:** [Aslam Doctor](https://profiles.wordpress.org/aslamdoctor/)
 * **Repository:** [GitHub](https://github.com/aslamdoctor/ai-provider-for-grok)·
   [WordPress.org](https://wordpress.org/plugins/aslams-ai-provider-for-grok/)
 * **Version:** 1.0.2
 * **Capabilities:** Text generation, function calling, vision input, structured
   output, multi-candidate generation, chat history
 * **Models:** Dynamically discovered from the xAI API, including vision-capable
   models
 * **API key:** Purchase API tokens from [x.ai](https://x.ai/) and configure under
   Settings > AI Credentials or via the `GROK_API_KEY` environment variable
 * **Install:** Install from [WordPress.org](https://wordpress.org/plugins/aslams-ai-provider-for-grok/)
   or download from [GitHub Releases](https://github.com/aslamdoctor/ai-provider-for-grok/releases)

### AI Provider for Mistral

[Mistral AI](https://mistral.ai/) is a European AI company offering powerful open
and commercial models for text and image generation.

 * **Author:** [Lauri Saarni](https://profiles.wordpress.org/laurisaarni/)
 * **Repository:** [GitHub](https://github.com/saarnilauri/ai-provider-for-mistral)·
   [WordPress.org](https://wordpress.org/plugins/ai-provider-for-mistral/)
 * **Version:** 1.1.0
 * **Capabilities:** Text generation, image generation, function calling, vision
   input, structured output (JSONJSON JSON, or JavaScript Object Notation, is a 
   minimal, readable format for structuring data. It is used primarily to transmit
   data between a server and web application, as an alternative to XML. schema)
 * **Models:** Dynamically discovered from the Mistral API — always up to date with
   the latest available models, including `mistral-medium-2505` for image generation
 * **API key:** Obtain from [Mistral Console](https://console.mistral.ai/)
 * **Install:** Install from [WordPress.org](https://wordpress.org/plugins/ai-provider-for-mistral/)
   or download from [GitHub Releases](https://github.com/saarnilauri/ai-provider-for-mistral/releases)

### AI Provider for Ollama

[Ollama](https://ollama.com/) lets you run open-source AI models locally on your
own hardware — no API key, no cloud, no cost. Perfect for privacy-conscious sites
and local development.

 * **Author:** [Fueled](https://profiles.wordpress.org/fueled/)
 * **Repository:** [GitHub](https://github.com/Fueled/ai-provider-for-ollama) · 
   [WordPress.org](https://wordpress.org/plugins/ai-provider-for-ollama/)
 * **Capabilities:** Text generation using locally hosted models, automatic model
   detection, support for both local and Ollama Cloud deployments
 * **Models:** Any model available through Ollama’s registry (Llama 3.2, Mistral,
   Phi, Gemma, and more)
 * **API key:** No API key required for local mode — just have Ollama running on
   your machine. Cloud mode requires an Ollama Cloud API key.
 * **Install:** Install from [WordPress.org](https://wordpress.org/plugins/ai-provider-for-ollama/)
   or download from [GitHub](https://github.com/Fueled/ai-provider-for-ollama). 
   For local mode, [install Ollama](https://ollama.com/download) and pull a model(
   e.g., `ollama pull llama3.2`).

### AI Provider for OpenRouter

[OpenRouter](https://openrouter.ai/) is a unified API gateway providing access to
400+ AI models from dozens of providers — including Anthropic, OpenAI, MetaMeta 
Meta is a term that refers to the inside workings of a group. For us, this is the
team that works on internal WordPress sites like WordCamp Central and Make WordPress.,
Google, Mistral, and many more — through a single API key.

 * **Author:** [Jonathan Bossenger](https://profiles.wordpress.org/psykro/)
 * **Repository:** [GitHub](https://github.com/jonathanbossenger/wp-openrouter-provider)·
   [WordPress.org](https://wordpress.org/plugins/ai-provider-for-openrouter/)
 * **Capabilities:** Text generation with access to 400+ models, searchable model
   selection with provider filtering, free-tier model toggle, context length and
   pricing display
 * **Models:** Full OpenRouter catalog — Claude, GPT-4, Llama 3, Gemini, Mistral,
   and hundreds more
 * **API key:** Create an account at [openrouter.ai](https://openrouter.ai/) and
   retrieve your key from Settings > Keys. Configure under Settings > OpenRouter
   AI in the WordPress admin.
 * **Install:** Install from [WordPress.org](https://wordpress.org/plugins/ai-provider-for-openrouter/)

## Testing environment

We encourage testers to use **WordPress 7.0** which includes the PHP AI Client as
part of core. This is the closest you can get to the final WordPress 7.0 experience
and helps us catch integration issues before the stable release.

### Prerequisites

You need a WordPress 7.0 environment to test these connectors. You have two
options:

#### Option 1 — Fastest: the experimental WP Dev Environment Toolkit (recommended for new testers)

The [experimental WP Dev Environment Toolkit](https://github.com/WordPress/experimental-wp-dev-env)

is a desktop app (macOS, Windows, Linux) that bundles GitGit Git is a free and open
source distributed version control system designed to handle everything from small
to very large projects with speed and efficiency. Git is easy to learn and has a
tiny footprint with lightning fast performance. Most modern plugin and theme development
is being done with this version control system. [https://git-scm.com/](https://git-scm.com/),
Node.js, and aPlayground-powered WordPress server. Install the app, pick a directory,
clicka button, and you have a running WordPress 7.0 dev site — no terminal, noDocker,
no manual dependency installs.

See the [announcement on make.wordpress.org/core](https://make.wordpress.org/core/2026/04/16/wordpress-core-dev-environment-toolkit-a-faster-path-to-your-first-core-contribution/)

for background. The tool is **experimental**; if you hit issues with theenvironment
itself (not the connectors), please file them on the[experimental-wp-dev-env repo](https://github.com/WordPress/experimental-wp-dev-env/issues).

#### Option 2 — Traditional local setup

If you already have a local WordPress dev environment (Local, Studio, wp-env,
Docker,
VVV, etc.), you can keep using it. You will need:

 * WordPress 7.0 (the [WordPress Beta Tester](https://wordpress.org/plugins/wordpress-beta-tester/)
   plugin is the easiest way to get it)
 * PHP 7.4 or higher (8.0+ for OpenRouter and Ollama)

#### In both cases, you also need

 * The **[AI Experiments](https://wordpress.org/plugins/ai/)** plugin installed 
   and activated, which provides the user-facing features that leverage the AI capabilities
   these connectors provide.
 * An API key for your chosen provider (not needed for local Ollama)

### Setup Instructions

#### Using the experimental WP Dev Environment Toolkit

 1. Download and install the toolkit from the
    [experimental-wp-dev-env releases page](https://github.com/WordPress/experimental-wp-dev-env/releases).
 2. Launch the app, pick a directory, and click the button to spin up a
    `wordpress-
    develop` checkout with a running dev server.
 3. In the WordPress admin, install the
    [WordPress Beta Tester](https://wordpress.org/plugins/wordpress-beta-tester/)
    plugin if you are not already on 7.0.
 4. Install and activate the **AI Experiments** plugin.
 5. Install one or more community connector plugins (links below).
 6. Go to **Settings → Connectors** and verify the provider appears in the
    connector
    selection screen.

#### Using your own local setup

 1. Ensure you are running WordPress 7.0 (use WordPress BetaBeta A pre-release of software
    that is given out to a large group of users to trial under real conditions. Beta
    versions have gone through alpha testing in-house and are generally fairly close
    in look, feel and function to the final product; however, design changes often 
    occur as part of the process. Tester if needed).
 2. Install and activate the **AI Experiments** plugin.
 3. Install one or more community connector plugins.
 4. Verify the provider appears in **Settings → Connectors**.

## What to test

### 1. Provider setup and connectivity

 * Does the connector plugin activate without errors?
 * Does the provider appear in the connector selection screen alongside the default
   providers (Anthropic, Google, OpenAI)?
 * Is the API key configuration straightforward?
 * Does the availability check succeed when the API key is valid?

### 2. Text generation

 * Enable AI experiments in the AI settings.
 * Open the post editor and and use text generation fetures. Do AI-powered suggestions
   appear?
 * Try switching between community connectors and the default providers — does the
   experience remain consistent?

![](https://make.wordpress.org/ai/files/2026/03/ai-experiments-enabled-1024x931.
jpg)

Screenshot of showing AI Editor Experiments being enabled.

### 3. Image generation

For connectors that support image generation (currently Mistral):

 * Can you generate images using “Generate featured imageFeatured image A featured
   image is the main image used on your blog archive page and is pulled when the
   post or page is shared on social media. The image can be used to display in widget
   areas on your site or in a summary list of posts.”- feature in the editor?
 * How does the generation speed compare to the default providers?

### 4. Error handling

 * What happens when an invalid API key is provided?
 * For Ollama: what happens when the local Ollama service is not running?
 * Are error messages clear and actionable?

## Known issues

 * Image generation is currently only available on select providers and models.

## Questions to consider

When testing, we’d appreciate your thoughts on:

 1. **Setup experience:** Was the installation and configuration process clear? What
    could be improved?
 2. **Provider switching:** Did switching between community connectors and default 
    providers feel seamless?
 3. **Response quality:** How did the AI responses compare across different community
    providers and the defaults?
 4. **Performance:** Did you notice latency or performance differences between providers?
 5. **Local AI (Ollama):** If you tested local models, how did the experience compare
    to cloud providers? Was setup straightforward?
 6. **Error messages:** When something went wrong, were the error messages helpful?
 7. **Documentation:** Was there enough information to get started? What was missing?

## How to give feedback

 * **Grok connector issues:** [GitHub Issues](https://github.com/aslamdoctor/ai-provider-for-grok/issues)
 * **Mistral connector issues:** [GitHub Issues](https://github.com/saarnilauri/ai-provider-for-mistral/issues)
 * **Ollama connector issues:** [GitHub Issues](https://github.com/jonathanbossenger/wp-ollama-model-provider/issues)
 * **OpenRouter connector issues:** [GitHub Issues](https://github.com/jonathanbossenger/wp-openrouter-provider/issues)
 * **AI Experiments issues:** [GitHub Issues](https://github.com/WordPress/ai-experiments/issues)
 * **PHP AI Client issues:** [GitHub Issues](https://github.com/WordPress/php-ai-client/issues)
 * **General discussion:** Join the [#core-ai](https://wordpress.slack.com/archives/core-ai)
   channel on [WordPress Slack](https://make.wordpress.org/chat/).

## Testing timeline

The original structured testing window ran **25 March – 8 April 2026**. Based on
community interest and due to the delayed release of [WordPress 7.0 release](https://make.wordpress.org/core/2026/04/22/wordpress-7-0-release-party-updated-schedule/),
we have extended open testing through **18 May 2026 **. Issues filed after that 
date are still welcome and will be triaged on the connector repositories listed 
in the article and it’s comments.

## Want to build your own connector?

The PHP AI Client makes it straightforward to build a connector for any AI service.
If your preferred AI provider isn’t listed above, consider building one! The [PHP AI Client documentation](https://github.com/WordPress/php-ai-client)
and the community connector plugins listed here are great starting points. At a 
high level, connector plugins need to:

 1. Extend `AbstractApiProvider` from the PHP AI Client.
 2. Implement model metadata discovery and model classes for supported capabilities.
 3. Register the provider with `AiClient::defaultRegistry()` on the WordPress `init`
    hook.

We welcome new community connectors and would love to include them in future testing
rounds.

---

Thank you for helping test and improve WordPress’s AI capabilities. The more providers
the community builds and tests, the more choice WordPress users will have when AI
features land in WordPress 7.0.

_Props to [@aslamdoctor](https://profiles.wordpress.org/aslamdoctor/), [@fueled](https://profiles.wordpress.org/fueled/),
[@laurisaarni](https://profiles.wordpress.org/laurisaarni/) and [@psykro](https://profiles.wordpress.org/psykro/)
for their contributions to the WordPress AI connector ecosystem._

_Thanks to [@jeffpaul](https://profiles.wordpress.org/jeffpaul/) for reviewing the
article._

**Updates**

_25 April 2026: Added the experimental WP Dev Environment Toolkit as the recommended
onboarding path. Clarified status of the testing window. Props to [@nikunj8866](https://profiles.wordpress.org/nikunj8866/)
for feedback. _

[#ai-experiments](https://make.wordpress.org/ai/tag/ai-experiments/), [#call-for-testing](https://make.wordpress.org/ai/tag/call-for-testing/),
[#core-ai](https://make.wordpress.org/ai/tag/core-ai/), [#needs-testing](https://make.wordpress.org/ai/tag/needs-testing/)

 [  ](https://profiles.wordpress.org/jeffpaul/) [Jeffrey Paul](https://profiles.wordpress.org/jeffpaul/)
1:27 am _on_ 4 February 2026     
Tags: [AI Experiments ( 16 )](https://make.wordpress.org/ai/tag/ai-experiments/),
call for testing, [core-ai ( 78 )](https://make.wordpress.org/ai/tag/core-ai/), 
[needs testing ( 2 )](https://make.wordpress.org/ai/tag/needs-testing/)   

# 󠀁[Call for Testing: Exploring New AI Experiments](https://make.wordpress.org/ai/2026/02/04/call-for-testing-exploring-new-ai-experiments/)󠁿

Over the past few weeks, we’ve been experimenting with a handful of AI-powered concepts
in the AI Experiments pluginAI Experiments Plugin WordPress's AI laboratory bringing
all building blocks together. Serves as both a user tool and developer reference
implementation. First release (v0.1.0) includes Title Generation experiment.. These
concepts are intentionally exploratory. Some are rough, some may not stick, and 
all of them need feedback from folks actually using WordPress in the real world.

This post is a Call for Testing, with less interest in “is this perfect?” and more
interested in:

 * Does this feel useful or confusing?
 * Would you expect this to exist where it does?
 * What would you change about the UXUX UX is an acronym for User Experience - the
   way the user uses the UI. Think ‘what they are doing’ and less about how they
   do it. or flow?
 * Does it behave as expected in your testing?

Below you’ll find a set of user-facing experiments, developer-focused tools, and
one hybrid feature that cuts across both. Each section includes a short description,
steps to test and a screencast example, a WordPress Playground link to utilize for
testing, and a link back to the PR for feedback.

Please test what you’re curious about. You don’t need to try everything.

## User-Facing Experiments

These experiments focus on AI features that surface directly in the WordPress admin
or content workflows.

### Type Ahead Suggestions

This experiment explores AI-powered type-ahead suggestions while writing. The goal
is to understand whether inline suggestions help with flow and clarity, or whether
they feel distracting or overly prescriptive.

We’re especially interested in feedback on:

 * Does the timing of suggestions feel right?
 * Do suggestions feel helpful or noisy?
 * Where would you expect controls or settings for this to live?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-151-850d097206bef23477edb240a29540029a6fff1e.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Credentials` and add an APIAPI An API or Application
    Programming Interface is a software intermediary that allows programs to interact
    with each other and share data in limited, clearly defined ways. key to one of 
    the AI clients listed, then click `Save Changes`
 4. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `Type-ahead Text`, and click `Save Changes`
 5. Navigate to `Posts > All Posts` and either edit the `Hello world!` post or click`
    Add Post` to open the post editor
 6. Start typing content and observe type-ahead suggestions
 7. Press `Tab` to accept or `Escape` to dismiss suggestions

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/151](https://github.com/WordPress/ai/pull/151)

---

### Comment Moderation Assistance

This experiment explores using AI to assist with comment moderation. Rather than
fully automating decisions, the focus is on providing helpful context or signals
to moderators.

We’re especially interested in feedback on:

 * How do you feel about the AI Reply and Analyze with AI functionality?
 * Does this feel like assistance or automation?
 * Is the information presented at the right moment?
 * What would increase or decrease trust here?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-155-0ee4cd8277ca18127d02d31c3350a8a31a9d4b1c.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Credentials` and add an API key to one of the AI clients
    listed, then click `Save Changes`
 4. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `Comment Moderation`, and click `Save Changes`
 5. Navigate to `Posts > All Posts`, view the `Hello world!` post, and leave some sample
    comments on the post
 6. Navigate to `Comments` and utilize the `AI Reply` row action on a sample comment,
    select a suggested reply to use, and then `Reply` to the comment
 7. Utilize the `Analyze with AI` option in the `Bulk actions` dropdown menu
 8. Review Sentiment and Toxicity scores for Comments

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/155](https://github.com/WordPress/ai/pull/155)

---

### Markdown Feeds

This experiment looks at consuming and working with Markdown-based feeds using AI.
The goal is to explore new content ingestion workflows and how AI might help interpret
or transform structured text.

We’re especially interested in feedback on:

 * Does this unlock workflows you care about?
 * Is Markdown the right abstraction here?
 * What would you expect next if this were expanded?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-194-461b4297c76cf5b03de5283504d4294911b8d16e.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `Markdown Feeds`, and click `Save Changes`
 4. Navigate to `Posts > All Posts` and view the `Hello world!` post
 5. Change the end of the URLURL A specific web address of a website or web page on
    the Internet, such as a website’s URL www.wordpress.org to `/hello-world.md` and
    confirm `text/markdown` output

Note additional testing options:

 * Visit `/?feed=markdown` and confirm `text/markdown` output
 * Request a post with `Accept: text/markdown` and confirm Markdown output
 * Optional: use the new section filters to inject custom fields into output: `npm
   run test:php -- --filter Markdown_Feeds`

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/194](https://github.com/WordPress/ai/pull/194)

---

### Extended AI Providers

This experiment expands the set of available AI providers, making it easier to test
and compare different models and services.

We’re especially interested in feedback on:

 * Is the providerProvider An AI service offering models for generation, embeddings,
   or other capabilities (e.g., Anthropic, Google, OpenAI). selection UX understandable?
 * Do the abstractions make sense from a user point of view?
 * What information would help you choose between providers?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-148-f99595c830bf60c57aa98ba67aacd45c18bf42ae.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Credentials` and add an API key to one of the three 
    default AI clients listed, then click `Save Changes`
 4. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `Extended Providers`, and click `Save Changes`
 5. Return to the `Settings > AL Credentials` screen and explore the additional nine
    providers and configuration options
 6. Add an API key for at least one cloud provider (e.g., Groq, DeepSeek) and verify
    it saves
 7. Try the same task across multiple providers

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/148](https://github.com/WordPress/ai/pull/148)

---

## Developer Experience Experiments

These experiments focus on developers building with, extending, or debugging AI 
features in WordPress.

### AI Playground

The AI Playground is a developer-focused space to experiment with prompts, providers,
and responses without wiring everything into a full feature.

We’re especially interested in feedback on:

 * Does this feel useful during development?
 * What’s missing that would make this part of your AI configuration process?
 * Are there any missing features or elements that you expected to find based on
   you understanding of “AI Playground”?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-140-caa26471697e01d35705a16266131d0e4ff564ae.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Credentials` and add an API key to one (**or more**)
    of the AI clients listed, then click `Save Changes`
 4. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `AI Playground`, and click `Save Changes`
 5. Navigate to `Tools > AI Playground` and test out different prompts, providers, 
    models, capabilities, and options
 6. Observe outputs from different provider, model, and option configurations

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/140](https://github.com/WordPress/ai/pull/140)

---

### MCP (Model Context Protocol) Integration

This experiment explores MCP (Model Context Protocol) support, aiming to improve
how context is shared and managed across AI interactions.

We’re especially interested in feedback on:

 * Does the MCP integration function as you would expect?
 * Is this solving a real pain point for you?
 * What documentation or examples would help adoption?

How to test:

 1. Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-152-0bb4efe08894bb8525f6ff123f717af30c894c2b.zip%22%7D%7D%5D%7D)
 2. Open the WP Admin dashboard
 3. Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
    Changes`, check `MCP`, and click `Save Changes`
 4. Navigate to the top level `MCP` menu item and view the MCP screen to ensure that`
    Enable MCP` and `Enable Server` are toggled on and that the server shows as `🟢
    Running`
 5. Toggle the `Expose via MCP` item for various abilities to test
 6. Use the `Test connection` feature in the `Connection test` section to verify endpoint
 7. Copy a `Client configuration` and test in Claude Desktop or Cursor

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/152](https://github.com/WordPress/ai/pull/152)

---

## Hybrid Experiment

### AI Request Logging

This experiment adds logging for AI requests (provider, model, tokens, duration,
cost estimate) that can be viewed from the Admin Dashboard as well as configurable
retention period and automatic cleanup to help with debugging, cost tracking, and
usage analysis.

We’re especially interested in feedback on:

 * Is the level of detail appropriate?
 * Who do you think this is for: site owners, developers, or both?
 * How should this data be surfaced or stored?
 * How would you want to filterFilter Filters are one of the two types of Hooks 
   [https://codex.wordpress.org/Plugin_API/Hooks](https://codex.wordpress.org/Plugin_API/Hooks).
   They provide a way for functions to modify data of other functions. They are 
   the counterpart to Actions. Unlike Actions, filters are meant to work in an isolated
   manner, and should never have side effects such as affecting global variables
   and output., hide, or export this data?

How to test:

 1.  Launch using [this WordPress Playground link](https://playground.wordpress.net/?blueprint-url=data:application/json,%7B%22steps%22%3A%5B%7B%22step%22%3A%22installPlugin%22%2C%22pluginZipFile%22%3A%7B%22resource%22%3A%22url%22%2C%22url%22%3A%22https%3A%2F%2Fgithub.com%2FWordPress%2Fai%2Freleases%2Fdownload%2Fci-artifacts%2Fpr-149-3731f8fc5ff9775350d7621fd68e5904ccb55d06.zip%22%7D%7D%5D%7D)
 2.  Open the WP Admin dashboard
 3.  Navigate to `Settings > AI Experiments`, check `Enable Experiments`, click `Save
     Changes`, check `AI Request Logging`, and click `Save Changes`
 4.  Enable other Experiments (and necessary, related AI Client Credentials via `Settings
     > AI Credentials`) and perform actions that trigger AI requests (e.g. Title Generation,
     ExcerptExcerpt An excerpt is the description of the blog post or page that will
     by default show on the blog archive page, in search results (SERPs), and on social
     media. With an SEO plugin, the excerpt may also be in that plugin’s metabox. Generation)
 5.  Navigate to `Settings > AI Request Logs` and view the AI Request Logs screen showing
     requests, tokens, and other data
 6.  Review logged requests and metadata
 7.  Test filtering by provider, status, and date range
 8.  Verify “Purge logs” clears all entries
 9.  Disable the experiment and confirm no new logs appear

Screencast:

Share feedback / leave comments: [https://github.com/WordPress/ai/pull/149](https://github.com/WordPress/ai/pull/149)

---

## How to Give Feedback

Feedback directly on the PRs is ideal, but broad impressions are also valuable (
e.g. as comments on this post, messages in [#core-ai](https://wordpress.slack.com/archives/C08TJ8BPULS)).
You don’t need to review code to participate, high-level product, UX, and architectural
feedback is just as valuable here. If you’re short on time, even answers to one 
or two of these questions help:

 * Would you use this if it shipped tomorrow?
 * What’s the first thing you’d change?
 * Where does this feel surprising, in a good or bad way?

These experiments are about learning in public. Your perspective can help us decide
what to refine, what to pause, and what to drop entirely. Thanks for helping shape
what AI in WordPress could become.

_Props to [@dkotter](https://profiles.wordpress.org/dkotter/) for helping with content
for this post and [@justlevine](https://profiles.wordpress.org/justlevine/) for 
pre-publish review._

[#ai-experiments](https://make.wordpress.org/ai/tag/ai-experiments/), [#call-for-testing](https://make.wordpress.org/ai/tag/call-for-testing/),
[#core-ai](https://make.wordpress.org/ai/tag/core-ai/), [#needs-testing](https://make.wordpress.org/ai/tag/needs-testing/)