Title: Specflux Site Agent
Author: Specflux Solutions
Published: <strong>October 14, 2026</strong>
Last modified: October 14, 2026

---

Search plugins

![](https://ps.w.org/specflux-site-agent/assets/banner-772x250.png?rev=3745120)

![](https://ps.w.org/specflux-site-agent/assets/icon-256x256.png?rev=3745120)

# Specflux Site Agent

 By [Specflux Solutions](https://profiles.wordpress.org/specflux/)

[Download](https://downloads.wordpress.org/plugin/specflux-site-agent.0.3.3.zip)

 * [Details](https://wordpress.org/plugins/specflux-site-agent/#description)
 * [Reviews](https://wordpress.org/plugins/specflux-site-agent/#reviews)
 *  [Installation](https://wordpress.org/plugins/specflux-site-agent/#installation)
 * [Development](https://wordpress.org/plugins/specflux-site-agent/#developers)

 [Support](https://wordpress.org/support/plugin/specflux-site-agent/)

## Description

Specflux Site Agent runs an agent loop on behalf of a logged-in user: one goal, 
many
 model turns, many tool calls — pausing for a human whenever a write needs 
approval, whenever the model needs to ask a clarifying question, or before its first
side-effecting write — and resuming exactly where it left off.

 * **Runs are session-bound.** A run acts as the user who started it; nothing
    runs
   in the background.
 * **Tools are WordPress Abilities**, restricted per run to an allow-list
    supplied
   directly by the consumer, or derived from a registered **pack** (a named tool
   surface — pages, posts, site, or commerce, all bundled).
 * **Two gate modes, resolved once per run.** If the **Belayline Agent Governance**
   plugin
    is active, every call is governed by its gate (packs, tiers, approvals)
   and audit trail. Without it, Specflux Site Agent falls back to a **built-in minimal
   gate: every call that changes the site still stops for a person to approve it
   on Specflux Site Agent’s own Runs screen before it runs; reads do not. A run’s
   mode is pinned at start and never changes underneath it — if the environment’s
   mode changes mid-run, the run fails cleanly instead of silently switching enforcement.
 * **Three kinds of park.** An **approval park** (a governed write, resolved
    by
   a person), a **question park** (the model asks the user one structured question),
   and a **plan park** (the model proposes a numbered plan before its first write;
   once accepted, that plan’s tool set is a fence — calls outside it are refused
   until a new plan is accepted).
 * **Budgets bound every run.** Steps, tool calls, tokens, questions and
    plans.
   A consumer may only lower a registered ceiling; running out of questions is not
   a failure — the model loses the ability to ask and proceeds on stated assumptions.
 * **Four bundled packs.** Pages and posts (create/update/publish, with
    publish
   handled as its own governed step), site (navigation, front page, homepage patterns),
   and commerce (WooCommerce catalogue and order operations, when WooCommerce is
   active). Every content write is validated before anything is persisted: unrenderable
   or non-round-tripping markup, blocks outside the allowed set, disallowed tags
   or attributes, unresolved {{placeholder}} text, an unrecognised pattern shape,
   a slot count out of range, a page-shape rule (e.g. hero must come first), or 
   a slug/title collision with an existing page or post — all refuse the whole write
   and persist nothing.
 * **Not for multisite.** Specflux Site Agent refuses to activate on a WordPress
   
   multisite install.

### External services

Specflux Site Agent itself makes two kinds of outbound request, both only while 
a
 human is actively driving a run from their own browser session — Specflux Site
Agent never runs in the background.

**Model calls (AI Client).** Made through the WordPress AI Client (bundled
 with
WordPress 7.0+), using whichever provider your site has connected under Settings
Connectors — commonly OpenAI, but any provider the AI Client supports. On every 
model turn (at most one per tick), Specflux Site Agent sends that provider:

 * the run’s full conversation history so far (the goal, every model
    message, and
   every tool result already produced in the run);
 * a system instruction assembled from the harness’s own operating rules,
    the active
   pack’s skills (e.g. a pack’s block-pattern vocabulary and copy constraints), 
   and anything the consumer or `specflux_site_agent_system_instruction` filter 
   contributes;
 * the declared tool (Ability) schemas the run is allowed to call, and the
    results
   those tools return, which are fed back to the model on the next turn (capped 
   at 32 KB per result by default, filterable via specflux_site_agent_tool_result_max_bytes).

No file, database or site content is sent beyond what a run’s own tool
 calls read
and return as results. Sending and handling this data is governed by the connected
provider’s own terms. The site owner chooses the AI provider, so that provider’s
terms and privacy policy apply; see, for example, OpenAI’s: https://openai.com/policies/
row-terms-of-use/ and https://openai.com/policies/row-privacy-policy/ — consult 
your specific provider’s terms if you have connected a different one.

**Stock photo search (Openverse).** When a run searches for a stock image
 (pages/
site packs), Specflux Site Agent sends the model-generated search text as a query
string to Openverse’s public search API (`https://api.openverse.org/v1/images/`)
and to Openverse’s individual image-detail endpoint when fetching a chosen result.
No account credentials, site content, or personal data are sent — only the search
terms and the id of a selected result. This runs only during an active, human-driven
run. See Openverse’s terms and privacy policy: https://docs.openverse.org/terms_of_service.
html and https://openverse.org/privacy (Openverse is a WordPress.org project; its
privacy policy is https://wordpress.org/about/privacy/).

**Image downloads.** When a run imports a chosen stock photo or a generated
 image,
the plugin downloads the image file from the URL Openverse or the AI provider returned(
Openverse results are hosted by third-party sites such as Flickr or Wikimedia Commons;
an AI provider may instead return the image data inline, in which case nothing is
downloaded). Only the image URL is requested, with no site data, and the file is
then saved to your Media Library.

### Privacy

Specflux Site Agent stores each run in two custom tables: the ID of the user who

started it, the goal they typed, the conversation with the AI model, the tool calls
and their results, and timestamps. Runs stay until you remove them or delete the
plugin with the opt-in below.

Runs are sent off-site to the AI provider you connected, and search terms
 are sent
to Openverse; see “External services” above for exactly what.

Specflux Site Agent adds suggested text for your privacy policy under Settings 

Privacy  Policy Guide. It does not register a personal-data exporter or eraser.

Deleting the plugin keeps your runs by default. To remove all Specflux Site Agent

data (both tables, its options, transients and per-user flags) when the plugin is
deleted, opt in first: `wp option update specflux_site_agent_uninstall_delete_data
1`, then delete the plugin.

### Developers

#### Extension API

    ```
    SPECFLUX_SITE_AGENT_API_VERSION (currently 0.3.0, defined in specflux-site-agent.php) versions the declared
    ```

extension surface, independently of this plugin’s own Stable tag. Semver promise:
a removal
 or signature change needs a major bump; an addition needs a minor bump—
this applies starting at 0.3.0, even below 1.0.

The `@api` list: the `Pack` abstract class’s `@api`-tagged methods (name, roles,

abilityNamespaces, inputProperties, verbFor, objectIdKey/Prefix/ForWrite/ForRead,
roleCapabilities, withheldRoleNotice, verbMap, roleVerbs, ungrantableVerbs, governedNamespaces,
agentSafetyVerbMap, defaultBudget, skills, agentSafetyPack, validateCall, setupChecks,
runCapability, requiresAgentSafety, agentSafetyBindingError, guidesHash); the Api\
LayoutVocabulary facade (names, sectionSchema, validate, imageUrls, rulesLines);
the Skill/SkillSource/SetupCheck value types a pack constructs; and the `specflux_site_agent_packs`,
specflux_site_agent_run_skills and `specflux_site_agent_default_budget` filters.
Every other filter, including specflux_site_agent_can_tick and `specflux_site_agent_http_consumers`,
is not part of this surface and may change without a version bump.

REST (`specflux-site-agent/v1`) is the public `@api` consumer surface; admin-ajax
is this plugin’s own private transport for its
 bundled Runs screen and is not guaranteed
to match REST’s shape.

Deprecation: a break goes through `_deprecated_hook()`/`_deprecated_function()` 
for at least
 one minor release before removal at the next major. None exist yet.

See the “Extension API” section of README.md in the public repository,
 https://
github.com/specflux/specflux-site-agent, for the full reference, including how to
regenerate the reflection snapshot (tests/Api/public-surface.json there) that enforces
this.

#### Source code and build

The JavaScript in `build/` is compiled from the sources in `assets/src/` in
 the
public repository, https://github.com/specflux/specflux-site-agent. To rebuild it:

 1. `git clone https://github.com/specflux/specflux-site-agent && cd specflux-site-
    agent`
 2. `npm ci`
 3. `npm run build`
 4. npm run build runs `wp-scripts build` and writes `build/`.

## Installation

 1. Install and activate Specflux Site Agent. It works standalone with a built-in
     
    approval gate.
 2. Optionally install and activate **Belayline Agent Governance** for its full audit
    
    trail, tiered approvals, and cross-plugin governance — Specflux Site Agent detects
    it automatically and switches gate modes on the next run started after activation.
 3. Consumers (e.g. marketing-analytics-chat) detect `specflux_site_agent()` and offer
    
    multi-step runs automatically.

## FAQ

### Can the AI add scripts or code to my site?

No. Specflux Site Agent never writes PHP, JavaScript or CSS files and never runs
code
 the model produces. A run only writes content: pages, posts, navigation, products
and store reports. Every piece of markup it writes is checked against a fixed list
of allowed blocks, tags and attributes before saving, and anything outside that 
list (a `<script>`, an inline event handler, an unknown block) refuses the whole
write. Specflux Site Agent does this itself rather than relying on WordPress’s HTML
filtering, because administrators are allowed unfiltered HTML.

### Does it work without Belayline Agent Governance?

Yes. Belayline Agent Governance is optional, not required. If it isn’t active, Specflux
Site Agent
 uses its own built-in minimal gate: every call that changes the site
still stops for a person to approve it on Specflux Site Agent’s Runs screen before
it runs; read-only calls do not. Installing Belayline Agent Governance later adds
its full audit trail, tiered approval thresholds, and governance shared across other
plugins — new runs pick it up automatically.

### Does a run keep going while I’m not looking?

No. Runs are session-bound and driven from the browser; nothing executes in
 the
background. A parked run simply waits until someone with permission resolves it.

### What happens to a run in progress when I upgrade from 0.2?

A run that was still parked or running at the moment of the upgrade fails
 on its
next tick with a partial report and the reason “started under 0.2” — 0.2’s `update-
post` tier changed under it, so it cannot safely continue. Completed 0.2 runs still
render their full history and report. Nothing is silently re-tiered or auto-resumed.

### Which post statuses can create-post and update-post write?

create-post never publishes — it refuses a `publish` or `future` status
 outright
and only ever creates a draft. `update-post` covers draft-state edits only; any 
transition to `publish`/`future`, or any edit to an already-public post, goes through
the separate `publish-post` ability, which is a higher-tier, separately governed
call.

### Will pages use my theme’s design?

On block themes, yes. Specflux Site Agent builds each section from one of your theme’s

own patterns when one fits the content, and otherwise uses its own layouts with 
your theme’s spacing and colours. It never adds colours of its own to a theme pattern.
To use only Specflux Site Agent’s layouts, untick “Use my theme’s block patterns”
under Specflux Site Agent  Settings. Developers can drop individual patterns with
the `specflux_site_agent_theme_patterns` filter.

### Does Specflux Site Agent support WordPress multisite?

No. Specflux Site Agent refuses to activate on a multisite install.

## Reviews

There are no reviews for this plugin.

## Contributors & Developers

“Specflux Site Agent” is open source software. The following people have contributed
to this plugin.

Contributors

 *   [ Specflux Solutions ](https://profiles.wordpress.org/specflux/)
 *   [ Stephen Paul Samynathan ](https://profiles.wordpress.org/stephen1204paul/)

[Translate “Specflux Site Agent” into your language.](https://translate.wordpress.org/projects/wp-plugins/specflux-site-agent)

### Interested in development?

[Browse the code](https://plugins.trac.wordpress.org/browser/specflux-site-agent/),
check out the [SVN repository](https://plugins.svn.wordpress.org/specflux-site-agent/),
or subscribe to the [development log](https://plugins.trac.wordpress.org/log/specflux-site-agent/)
by [RSS](https://plugins.trac.wordpress.org/log/specflux-site-agent/?limit=100&mode=stop_on_copy&format=rss).

## Changelog

#### 0.3.3

Changed:

 * New name: Specflux Site Agent. Same features; new slug, text domain and identifiers.

#### 0.3.2

Fixed:

 * The bundled image patterns no longer reference a remote placeholder image; they
   use a small image shipped with the plugin.
 * Media search and the missing-alt-text list now need the upload_files capability
   and only show images the current user may read or edit. Saving a store report,
   setting a featured or product image, and enabling a coupon check the matching
   post type and attachment capabilities too.
 * A question the assistant asks is no longer cut off mid-word at 300 characters.
 * A plan step now shows one plain-language badge (Read only, Makes changes, Needs
   approval) with the actions it will use, instead of the same badge once per action.
 * A blank assistant turn no longer shows as an empty chat bubble, and a run still
   working in the background no longer greys out the Accept and Approve buttons 
   on another run.

#### 0.3.1

Fixed:

 * Sanitize the decoded JSON bodies of the admin-ajax start (budget) and tick
    (
   park resolution) requests: answer text, choices and plan notes go through sanitize_textarea_field(),
   budget caps are kept only as non-negative integers.
 * Point the Openverse terms-of-service link at its current location.

#### 0.3.0

New:

 * Two gate modes: Belayline Agent Governance when active, otherwise a built-in 
   minimal
    gate that still pauses every governed write for human approval on the
   Runs screen. A run’s mode is pinned at start.
 * Three new packs alongside pages: posts, site (navigation, front page,
    homepage
   patterns), and commerce (WooCommerce catalogue/order operations).
 * Stock photo search and alt-text/image-generation abilities via the
    WordPress
   AI Client and Openverse.
 * The Runs screen moved out of Tools to its own top-level “Specflux Site Agent”
   menu
    and is now a React app (`assets/src/runs/`, built with @wordpress/scripts);`
   tools.php?page=specflux-site-agent-runs` redirects there.
 * Extension API: `SPECFLUX_SITE_AGENT_API_VERSION`, a reflection-enforced `@api`
   
   surface for third-party packs (see “Extension API” above).
 * Pages follow your theme. Each page layout (hero, text, text with image,
    services,
   FAQ, call to action) is built from the active theme’s own block patterns where
   one fits, filled with the run’s copy. Twenty Twenty-Five and Ollie have built-
   in mappings; other block themes are matched automatically, and anything that 
   doesn’t fit falls back to Specflux Site Agent’s own layouts, using the theme’s
   spacing and colours.
 * Section tone: a section can sit on a contrast or accent band taken from
    the 
   theme palette, only where the text keeps WCAG AA contrast.
 * A Settings checkbox, “Use my theme’s block patterns”, to build every page
    from
   Specflux Site Agent’s own layouts instead.

**Breaking changes from 0.2:**

 * `specflux/update-post` is now draft-state edits only (Tier 1) — it
    refuses an
   already-public target (publish/future/private) and a requested publish/future
   transition. New `specflux/publish-post` (Tier 2) covers every publish/future 
   transition and any edit to an already-public post. Update your integration to
   call `publish-post` for those cases.
 * `specflux/create-post` now refuses `slug_collision` (409) when a
    non-trashed
   page or post of the same type already holds the requested slug, or matches the
   title case-insensitively.
 * Belayline Agent Governance is no longer a hard requirement. 0.2 always failed
   closed
    without it; 0.3 falls back to a built-in minimal gate instead (see the
   FAQ above). If you relied on `specflux_site_agent_ungoverned` meaning “Belayline
   Agent Governance is missing,” check the run’s `gate_mode` instead.
 * A run still parked or running under 0.2 at the moment you upgrade to 0.3
    fails
   on its next tick with reason “started under 0.2” instead of continuing — it is
   not migrated or resumed.
 * `GET /runs` (and the Runs screen’s run list) is now scoped to what the
    current
   viewer may see or resolve, instead of returning every run on the site to any 
   user who could reach the endpoint.

#### 0.2.0

 * Schema v2: run status/step-kind enums, `resume` object replaces the
    0.1 `approval_action`
   parameter (breaking), `max_questions`/`max_plans` budget keys.
 * Default budget ceilings raised to `max_steps` 60, `max_tool_calls` 30 and
    max_tokens
   250000 — a live pages run costs 31-45 steps and up to 90k tokens once the model
   retries a refused write, and a consumer may only lower a ceiling.
 * Question parks (`specflux/ask-user`) and plan parks
    (`specflux/propose-plan`)
   with an accept/veto fence around Tier ≥ 1 writes.
 * Packs: a named tool surface a run may start from (`pack` argument to
    start()),
   plus the bundled pages pack and its five polyfill Abilities.
 * Skills and a rendered system-instruction pipeline
    (`specflux_site_agent_run_skills`
   filter), with a per-run disable list (`skills_disable`).
 * Verification read-back and a per-run report surfaced on completion.
 * Runs screen: new-run form with preflight, three park cards, polling
    detail view.

#### 0.1.0

 * Initial scaffold: hard-dependency gate against Agent Safety.

## Meta

 *  Version **0.3.3**
 *  Last updated **16 hours ago**
 *  Active installations **Fewer than 10**
 *  WordPress version ** 7.0 or higher **
 *  Tested up to **7.1.3**
 *  PHP version ** 8.1 or higher **
 * Tags
 * [Agents](https://wordpress.org/plugins/tags/agents/)[AI](https://wordpress.org/plugins/tags/ai/)
   [approvals](https://wordpress.org/plugins/tags/approvals/)[automation](https://wordpress.org/plugins/tags/automation/)
   [safety](https://wordpress.org/plugins/tags/safety/)
 *  [Advanced View](https://wordpress.org/plugins/specflux-site-agent/advanced/)

## Ratings

No reviews have been submitted yet.

[Your review](https://wordpress.org/support/plugin/specflux-site-agent/reviews/#new-post)

[See all reviews](https://wordpress.org/support/plugin/specflux-site-agent/reviews/)

## Contributors

 *   [ Specflux Solutions ](https://profiles.wordpress.org/specflux/)
 *   [ Stephen Paul Samynathan ](https://profiles.wordpress.org/stephen1204paul/)

## Support

Got something to say? Need help?

 [View support forum](https://wordpress.org/support/plugin/specflux-site-agent/)