stable key, name, category
Let an AI agent prepare a change for review
Agent Bridge is the governed AI connection for Capell. It exposes the same capability layer used by Capell's AI Command Centre through MCP: an external AI client can discover intents, examples, and supported command shapes, validate structured inputs, find pages and sites by human-readable references, and request registered operations without receiving broad database or admin access.
Discoverable schemas
Scoped authority
Preview before mutation
Turn a brief into something reviewable
Prepare a private site preview, then decide what deserves to become a real change.
AI-assisted building is most useful when it creates a bounded result your team can inspect. Capell's AI Creator workflow connects a brief to a private preview and its approval path. It is an optional package, not a substitute for deciding the content, application behaviour or release acceptance.
Agent Bridge exposes a typed catalogue of Capell operations through MCP. The external client can discover capability names, categories, descriptions, intents, examples, input and output schemas, risks, scopes, package requirements, preview support, confirmation rules, and audit events before choosing a tool.
The Capell Agent Schema v1 draft adds a versioned contract for typed property sets, schema.org JSON-LD, declarative manifests, and the site-scoped /agent/v1 read API. Its Phase 2 implementation is under release verification; the supporting release must ship before this is presented as available on customer sites. Public results remain limited to published, agent-visible data.
The tokenless public endpoint is limited to allowlisted AI Creator catalogues, interview, schema, and validation. The authenticated endpoint can list, run, and confirm site capabilities within token scope and current policy. A mutation receives a non-destructive preview, a separate matching confirmation, another policy check, and a secret-scrubbed audit record.
This external-agent path runs no inference in Capell. Optional AI Creator inside the admin is a separate provider-backed workflow. Both paths share structured site validation and keep generated sites private until a person confirms the exact application plan.
Write a brief with a purpose
Give the preview a reader, a task and a clear boundary.
Describe the audience, the page family and the action a visitor should take. Supply approved facts and identify what must not change. A vague instruction to improve everything makes it harder to judge whether the result is useful or within scope.
Review the supported package workflow and installed capability set before starting. Keep application secrets out of the brief and distinguish a private preview from published content. Use existing page patterns where they fit instead of treating every request as a new website.
Approve the actual result
Review content, navigation and behaviour before accepting a change.
Inspect the preview on a small screen as well as a desktop. Check factual claims, image relevance, headings and the intended action. A coherent-looking page can still omit the information a visitor needs to decide.
Follow the package's confirmation and adoption process, keeping consequential publication or application work under its normal review. Record what was accepted and retain the recovery path. AI assistance does not make an unsupported capability available.
Before you adopt the AI packages
Both packages are paid; each listing states its current release and whether checkout is open. Use the live demo to inspect the public workflow first, then test the package path away from production. Keep a human responsible for provider terms, review, confirmation, and the final production release.
- Stable
- Agent Bridge
- Paid package with a stable current release: £39.00 for 12 months; its listing states the current release and whether checkout is open.
- Beta
- AI Creator
- Paid package with a beta current release; its listing shows when checkout opens.
- Review first
- Production
- Pilot away from production and keep a person accountable for approval.
- Live demo
- Try it
- Inspect the public command flow before choosing a package.
A maturity label does not grant installation access. Buy or confirm access on each listing, and use the demo to assess the workflow before committing to it.
How an agent request becomes a Capell operation
Discover
List visible capabilities with names, descriptions, categories, intents, examples, schemas, risk, and confirmation rules.
Suggest
Match the user's words to likely capabilities without treating a suggestion as permission to execute.
Resolve
Find an accessible page, site, or package from its title, slug, URL, domain, name, or registered key.
Authorise
Check token scopes, current user policy, site access, installed package state, and native workflow readiness.
Preview
Return the exact non-mutating impact, warnings, target, and expected result for review.
Confirm
Bind approval to one user, capability, payload, preview, expiry, and audit event, then allow one execution.
Capell Agent Schema v1
A versioned contract for agent-readable content
Capell Agent Schema v1 brings one documented shape to typed property sets, schema.org JSON-LD, declarative tool manifests, and the /agent/v1 read API. The contract follows the same site, language, domain, and publication rules as anonymous page delivery, so drafts and private editor data stay out of the public result.
Read the Capell Agent Schema v1 draft. The link becomes the canonical public specification when the supporting release is shipped; this page does not imply that the current uncommitted implementation is live on every Capell site.
Public and authenticated endpoints
- Public read,
ai/capell/read: list the allowlisted theme, page-type, section-type, layout, interview, schema, and validation capabilities. It has no preview, export, build, apply, tokens, private records, page administration, or package activation. - Authenticated,
ai/capell: list site capabilities and run or confirm those allowed by the Agent Bridge token, current policy, package state, and site context. - A valid input is not authority. Every capability carries its own risk and confirmation contract, and policy is checked again immediately before execution.
Inspect capability details before invocation
Capability discovery returns enough metadata for autocomplete and agent planning while keeping package internals private, including the direct-command shapes each capability accepts. Core page, site, publishing, package, and system operations live beside extension-contributed capabilities such as AI Creator. Installing or disabling an extension changes the catalogue and its command shapes without patching Agent Bridge.
description, intents, examples
input and output schemas
scope, server, policy ability
risk, preview, confirmation
required package and audit event
Decide what the agent may do
Does the task have a named target and typed result?
- Proceed with Agent Bridge when
- The capability can resolve one permitted site, page, package, or operation and validate a bounded output schema.
- Keep the current workflow when
- The request depends on broad database access, arbitrary HTTP, a shell, or an open-ended instruction.
Does the task need write authority?
- Proceed with Agent Bridge when
- Read discovery is insufficient and the smallest required mutation scope can be named explicitly.
- Keep the current workflow when
- The public read endpoint or an existing Filament action already completes the job.
Can a person inspect the exact impact before execution?
- Proceed with Agent Bridge when
- The capability produces a non-destructive preview and binds one-use confirmation to the same user, payload, and operation.
- Keep the current workflow when
- The proposed effect cannot be made legible or policy must be bypassed to complete it.
Can a named operator own the audit and revoke access?
- Proceed with Agent Bridge when
- One owner can review secret-scrubbed events, rotate or revoke the token, and investigate rejected operations.
- Keep the current workflow when
- No person owns scopes, confirmation, incident review, or the external model and retention terms.
Website-lifecycle tools without blanket admin access
Start with read capabilities, then grant only the write scopes needed for the delivery workflow. A token can discover a broader catalogue than it may execute, and the capability still applies current user and site policy.
Questions
Before you issue an agent token
Does Capell choose or run the external model?
No. The MCP and Agent Bridge path runs no inference. The external client supplies the reasoning, credentials, model choice, usage cost, and retention terms. Optional in-product AI Creator is separate and uses the provider your team configures through the AIOrchestrator package.
Can an agent copy another website?
No. Page cloning is limited to content already inside the authorised Capell site. AI Creator may use abstract reference notes, but its prompt and validation rules reject scraping, close reproduction, impersonation, third-party branding, logos, and copied trade dress.
Can a voice request install code from the internet?
No. Package activation is limited to Composer-present, registered Capell packages. The preview shows declared impact before confirmation. Agent Bridge does not offer arbitrary shell or Composer execution.
Does a successful preview guarantee the operation will run?
No. The preview can reject an unready request. A successful mutation preview still needs the matching confirmation, and policy is checked again at execution. Tokens expire, are bound to the user and payload, and can be used once.
What reaches the public website?
Only approved render data. Prompts, token scopes, permissions, internal record identifiers, field paths, audit entries, signed editor URLs, model metadata, and private previews stay outside public output.
Loading section
Next step
Start with read access
Start with discovery, then add each additional authority deliberately. Connect the public read endpoint first. Add a narrow token only after the team can explain its scopes, previews, confirmations, and audit review.
First task on the public read endpoint, with no token: on the site you are evaluating, open its ai/capell/read address (the path is on that site's own origin and can be changed per deployment, so confirm it with whoever runs the site). Call capell-public-list-capabilities to see what the allowlist exposes, run capell.ai-creator.interview.get for the question script, then check one small draft site spec with capell.ai-creator.discovery.validate_spec and read the errors it returns.
The named-page task is separate and protected: finding a page, previewing a private clone and confirming an operation need an installation configured for Agent Bridge, an authorised token and the authenticated ai/capell endpoint. None of that is available on the public read endpoint, and each package installs only with access from its own Marketplace listing. The in-admin AI-ready page shows how the same capability layer works for editors and voice input.