How to Build a WordPress Permission Test Matrix for AI Agents

A permission matrix should prove both allowed and refused WordPress actions for each AI identity, not merely list intended roles or demonstrate one successful request.

AI is most useful here as an evidence organizer, comparison engine and drafting assistant. It can make a complex WordPress task easier to inspect, but it cannot create missing authority, certify facts it did not observe or silently convert a recommendation into permission to act.

In one sentence: A permission matrix should prove both allowed and refused WordPress actions for each AI identity, not merely list intended roles or demonstrate one successful request.

What this guide helps you accomplish

Build an executable matrix that connects identities, capabilities, objects, states and expected outcomes to reproducible positive and negative evidence.

• An identity-by-action permission matrix with exact expected responses.
• Positive, negative, object-ownership and state-transition fixtures.
• A distinction between authentication failure, authorization denial, validation failure and unsupported capability.
• A regression suite for protected modes and custom abilities.

The finished artifact should be understandable by the person responsible for the decision and reproducible by someone who did not participate in the original prompt. A fluent answer is not enough. Every material conclusion needs a source, a scope and a verification path. When the evidence cannot establish something, the correct output is an explicit unknown or a testable hypothesis.

Evidence and inputs to prepare

• The current role, capability and WP Agent Control coverage contracts.
• Registered REST routes or abilities and their permission callbacks.
• Dedicated test users and safe WordPress fixtures.
• Expected HTTP, tool or application-level outcomes.
• A clean environment reset and evidence-retention plan.

Before supplying evidence to an assistant, remove credentials, secret values and unrelated personal information. Preserve the identifiers, versions, timestamps, locale, units and source labels needed to interpret what remains. A screenshot without a URL, state or date may be useful context, but it is rarely sufficient authority for a production decision.

Do not begin with a broad request such as “review this,” “fix this” or “make it better.” Define the decision the work must support, the population included, the source that is authoritative for each field, the allowed operations and the actions that remain forbidden. The planning or research stage should use a local repository, isolated fixture or exported evidence and does not require production WordPress access.

Role intention is not permission evidence

The matrix must exercise the actual endpoint, ability or WordPress action in the relevant object state.

A denial needs classification

401, 403, validation errors and unsupported operations have different meanings. Recording only failed hides the control actually tested.

Object and state matter

An identity may edit its own draft but not another author’s post, or update a draft but not a published page. Test the relevant boundaries.

Keep observation, inference and authority separate

A controlled review should distinguish at least four states:

1. Observed: directly present in a named record, file, response, rendered page or executed test.
2. Inferred: a plausible interpretation supported by evidence but not directly established.
3. Recommended: a proposed human decision or next action.
4. Authorized and verified: a separately approved change that was executed and then checked against acceptance criteria.

AI output usually begins in the first three states. It does not become authorized merely because it is detailed, internally consistent or technically convincing. Preserve this distinction in tables, reports, tickets and public case studies.

A safe workflow

1. Inventory identities, profiles, routes, abilities, objects and states.
2. Define the expected allow or deny outcome and rationale for every material cell.
3. Create isolated fixtures with stable IDs and reset procedures.
4. Execute allowed cases and retain exact request and result evidence.
5. Execute forbidden, malformed and out-of-scope cases.
6. Investigate every mismatch without widening permissions to make the test pass.
7. Add verified cases to automated regression coverage where practical.
8. Publish only a sanitized matrix and revoke temporary identities.

This sequence deliberately places accountable review between analysis and implementation. If a later stage needs broader access, create a new task, a new identity or an explicit permission change. Do not quietly upgrade the analytical identity because it reached a correct boundary.

Prompt recipe

Replace every value in square brackets before using the prompt. Do not paste passwords, API keys, authentication cookies, private customer records or unrelated personal information.

You are reviewing [TASK SCOPE] for [SITE, REPOSITORY OR DATASET] using only the supplied evidence.

Objective:
Build an executable matrix that connects identities, capabilities, objects, states and expected outcomes to reproducible positive and negative evidence.

Return the following fields:
- Identity
- Profile
- Object
- State
- Action
- Route or ability
- Expected outcome
- Expected status
- Observed outcome
- Evidence
- Disposition
- Version

Rules:
1. Use dedicated test identities and non-production fixtures.
2. Test both allowed and forbidden actions.
3. Preserve exact response classes and error bodies after sanitization.
4. Do not reinterpret an unexpected denial as permission to grant more access.
5. Do not publish credentials or sensitive endpoint details.

For every finding:
- identify the exact source, record, URL, file, line, object ID, state or dataset row;
- preserve dates, versions, units, locale, identifiers and denominators;
- separate observation, inference, recommendation and unknown;
- state what evidence was not available;
- do not change WordPress, source code, commerce data, analytics, external systems or published content.

Why this prompt is structured this way

The prompt creates an evidence contract before asking for recommendations. It makes missing data visible, reduces the chance that a model will complete an incomplete record with plausible prose and produces an output that can be reviewed systematically. Structured fields also make it easier to compare repeated runs or hand an approved subset to a later implementation workflow.

A production implementation may add JSON schema, typed tool inputs or automated validation. Those mechanisms improve consistency, but they do not establish that the source evidence is true, complete or current. Human review and system-specific verification remain required.

Recommended access boundary

Use No WordPress access during the planning or research stage for the stage described in this guide. The exact capabilities available to an identity must come from the installed product version, the published coverage contract and the connection method actually in use.

What must remain outside this task

• Permission changes
• Full Power fallback
• Production tests
• Expected-result rewriting after execution
• Unsupported security guarantees

A refused action can be useful evidence that the control boundary is working. Do not respond to an expected refusal by granting a broad administrator account or Full Power. First determine whether the action belongs in the current mandate at all. If it does, create a separately authorized stage with the narrowest required capability.

How WP Agent Control fits

This is a general WordPress workflow, not a promise that Agent Control can edit every object or integration discussed here. For the guided path, start with public pages; plugin, theme, user, setting, file, deletion, WooCommerce, ACF and builder operations are not native guided tasks. Use separately qualified tools and permissions where required.

Get structured site information and inspect selected published pages after connecting. No temporary task is needed for this public reading. You can also browse public pages without the plugin; Agent Control adds structured access and a path toward authorized WordPress work.

Connect your AI: docs first profile · See features and compatibility: coverage

Verification checklist

• ☐ The task, population, period, environment and decision are explicit.
• ☐ Every material observation is linked to exact evidence or labelled as a hypothesis.
• ☐ Stable IDs, URLs, versions, dates, units, locales and denominators are preserved.
• ☐ Missing evidence and coverage limits remain visible.
• ☐ The analytical or research identity performed no prohibited mutation.
• ☐ A qualified owner reviewed security, accessibility, legal, commerce or release implications where applicable.
• ☐ Any implementation has a separate mandate, access level, backup and verification plan.
• ☐ Temporary identities, fixtures and sensitive evidence are revoked, reset or disposed of after the task.

Common failure modes

• Success-only demonstration: The matrix proves that an action works but never that prohibited actions fail.
• Role-name proxy: Expected outcomes are copied from role labels without testing filtered or custom capabilities.
• Fixture contamination: One test changes the object state and invalidates later results.
• Denial flattening: All failures are treated as equivalent, hiding authentication or validation defects.

A recurring cross-cutting failure is permission drift: the initial task encounters a limit, and the operator broadens access before determining whether the missing operation is necessary, supported or safe. This destroys the evidence value of the refusal and makes later results difficult to attribute.

Advanced note

A governed permission matrix can be generated from the declared authority graph, but the declaration remains only an expectation until executed evidence confirms every material cell. Unexpected allowances are defects; expected denials are product proof.

Related guides

• Which WordPress Access Level Should You Give an AI?
• WordPress Abilities API Guide for AI Workflows
• How to Expose a Custom WordPress Ability through MCP
• How to Build a WordPress Test Plan with AI

Next step

Continue with the most relevant supporting guide and use the access-level guide before any authenticated task. When temporary WordPress access is no longer needed, finish by revoking the identity.