More screenshots
Works with the AI tools you already use
Admin Panel UI Architect
It begins by defining the authorized operators, their responsibilities, task frequency, access scope, operational risk, data sensitivity, and escalation paths.
$7.99
Admin Panel UI Architect
Example session with this skill installed
Product name
NexusFlow B2B Platform
Product type
Multi-tenant SaaS workflow automation platform
Admin audiences
Platform administrators
Support agents
Billing operators
Customer-success managers
Security reviewers
Read-only auditors
Organization administrators
Operator roles
Super Administrator
Platform Operator
Support Agent
Billing Administrator
Security Reviewer
Organization Owner
Workspace Administrator
Read-Only Auditor
Primary operational jobs
Create and manage organizations
Invite and manage users
Assign roles
Review account status
Investigate failed workflows
Manage subscriptions
Review invoices
Resolve payment failures
Review usage
Handle support escalations
Review audit history
Manage integrations
Suspend compromised accounts
Export approved operational reports
Tenant model
Platform
Organization
Workspace
Project
Entities
Users
Organizations
Workspaces
Projects
Roles
Permissions
Invitations
Sessions
Subscriptions
Plans
Invoices
Payments
Credits
Usage Records
Workflows
Workflow Runs
Integrations
API Keys
Support Tickets
Audit Events
Notifications
Reports
Account states
Invited
Pending Verification
Active
Suspended
Locked
Disabled
Archived
Deleted
Roles
Require verification against the actual authorization model
Permissions
View users
Invite users
Edit users
Suspend users
Delete users
Manage roles
Manage billing
View audit logs
Export data
Impersonate users
Manage integrations
Manage API keys
Issue refunds
Exact permission identifiers and inheritance require verification
Permission inheritance
Platform roles may grant organization-level access
Organization roles may grant workspace-level access
Exact inheritance rules require verification
Data sensitivity
Personal user data
Billing records
Payment status
API-key metadata
Authentication history
Support notes
Audit events
Workflow input and output metadata
User-management requirements
Search
Filter by status, role, organization, and activity
Invite users
Resend invitations
Change role
Suspend and restore
Revoke sessions
Review organization membership
View audit history
Archive and delete according to verified policy
Organization-management requirements
Create organization
Assign owner
Manage workspaces
View subscription
Review usage
Review billing
Manage domains
Manage integrations
Suspend organization
Transfer ownership
Billing requirements
View subscriptions
Change verified plans
View invoices
Review payment failures
Manage approved credits
Issue authorized refunds
Review usage
Cancel or reactivate subscriptions
All financial rules require verification
Audit requirements
Record actor
Action
Object
Scope
Timestamp
Result
Source
Before and after values where appropriate
Request or trace identifier
Retention rules require verification
Reporting requirements
User growth
Organization growth
Subscription status
Revenue summary
Usage
Failed workflow runs
Support volume
Security events
Moderation requirements
Not required in the first release
Support requirements
Customer overview
Organization context
Subscription
Usage
Recent failures
Tickets
Internal notes
Controlled impersonation
Escalation
Impersonation requirements
Explicit permission
Required reason
Time limit
Visible banner
Restricted actions
Audit event
Clear exit
Customer notification requirements need verification
Bulk actions
Assign role
Suspend users
Archive users
Export selected records
Apply tags
Move users between approved groups
Search requirements
Search by user name
Email
Organization
Workspace
Subscription identifier
Invoice number
Workflow-run identifier
Exact identifiers
Cross-tenant access must be permission controlled
Export requirements
CSV export for authorized reports
Audit export
User export
Billing export
Export permissions and masking require verification
Frontend stack
Next.js
React
TypeScript
Tailwind CSS
TanStack Table
Radix UI
Backend constraints
REST and event-driven services
Some actions complete asynchronously
Cursor-based pagination
Server-side authorization
Audit events may be delayed
Partial bulk-action failures are possible
Design system
Custom design system based on shadcn/ui and Radix UI
Responsive targets
Wide desktop
Desktop
Laptop
Tablet
Mobile for urgent operational tasks
Accessibility target
WCAG 2.2 AA implementation review
Localization
English initially
German and Spanish planned
Allow at least 30% text expansion
Need output
Complete admin information architecture
Operator personas
Navigation
Tenant and environment context
Global search
Dashboard
User management
Organization management
Roles and permissions
Permission matrix
Data tables
Filters
Saved views
Bulk actions
Detail pages
Settings
Audit logs
Billing
Reports
Log viewer
Support console
Impersonation
Destructive actions
Responsive behavior
Accessibility
Analytics
Developer handoff
Acceptance criteria
Regression tests
QA checklist
Risk assessment
Need variations
Operations-Focused Admin
Support-Focused Admin
Enterprise Governance Admin
Special constraints
Do not invent authorization rules.
Do not invent billing behavior.
Do not expose cross-tenant data.
Do not rely on hidden controls for authorization.
Do not allow silent impersonation.
Do not display private production data in public mockups.
Clearly distinguish synchronous and asynchronous actions.
Support partial bulk-action failures.
- Read your context and instructions
- Compiled the admin panel ui
- Generated the document
=== ADMIN PANEL UI ARCHITECTURE ===
Product
NexusFlow B2B Platform
Product Type
Multi-tenant SaaS workflow automation platform
Primary Operators
Platform administrators, support agents, billing administrators, security reviewers, organization administrators, and read-only auditors.
=== EXECUTIVE SUMMARY ===
NexusFlow requires an operations-focused administration system that keeps tenant, workspace, environment, authorization, and action risk visible throughout the interface.
The recommended architecture should be organized around operator domains rather than backend services:
- Overview
- Organizations
- Users
- Workflows
- Billing
- Support
- Audit Logs
- Reports
- Settings
The system should include persistent organization context, global entity search, role-aware navigation, structured detail pages, reusable table and filter foundations, safe bulk actions, human-readable audit history, and strong safeguards for impersonation and destructive operations.
=== OPERATOR PERSONAS ===
Operator
Platform Administrator
Responsibilities
Manage platform-wide organizations, users, roles, integrations, and high-risk system settings.
Risk
Critical
Data Access
Platform-wide, subject to verified authorization.
Operator
Support Agent
Responsibilities
Investigate customer problems, review account context, inspect workflow failures, and guide recovery.
Risk
High
Restrictions
No billing changes, permission changes, refunds, or destructive actions unless explicitly granted.
Operator
Billing Administrator
Responsibilities
Review subscriptions, invoices, payments, credits, usage, approved refunds, and billing failures.
Risk
High
Operator
Read-Only Auditor
Responsibilities
Review approved reports, audit events, account history, and selected security records.
Risk
Medium
Restriction
No modifying actions.
=== SCOPE MODEL ===
Scopes
- Platform
- Organization
- Workspace
- Project
Persistent Scope Indicator
Display the active organization and workspace in the application shell.
Environment Indicator
Show Production, Test, or other verified environment clearly.
High-Risk Rule
Destructive and billing actions must repeat the active scope inside the confirmation.
Cross-Tenant Rule
Search results, lists, exports, and actions must respect server-side tenant authorization.
=== PRIMARY NAVIGATION ===
Overview
Operational status and priority queues
Organizations
Organization management, ownership, workspaces, plans, usage, and account status
Users
Users, invitations, roles, sessions, status, and membership
Workflows
Workflow definitions, runs, failures, and integrations
Billing
Subscriptions, invoices, payments, credits, refunds, and usage
Support
Customer context, tickets, notes, recent failures, and controlled impersonation
Audit Logs
Human-readable administrative and security events
Reports
Approved operational and billing reports
Settings
Platform, organization, security, integration, notification, and developer settings
=== GLOBAL SEARCH ===
Searchable Objects
- Users
- Organizations
- Workspaces
- Subscriptions
- Invoices
- Workflow Runs
- Support Tickets
Search Result Anatomy
- Entity type
- Primary label
- Secondary identifier
- Organization context
- Status
- Match reason
- Authorized action
Rules
- Support exact identifiers.
- Mask sensitive fields.
- Display tenant context.
- Prevent cross-tenant disclosure.
- Support keyboard navigation.
- Provide no-results and permission-restricted states.
=== OVERVIEW DASHBOARD ===
Priority Modules
- Failed workflow runs
- Suspended organizations
- Payment failures
- Pending support escalations
- High-risk permission changes
- Recent administrative actions
- Usage anomalies
Each module should answer
- What requires attention?
- Why?
- Which entity is affected?
- What action is available?
- Which permission is required?
Avoid decorative metrics without operational value.
=== USER LIST ===
Primary Columns
- User
- Status
- Organization
- Role
- Last Active
- Authentication Method
- Actions
Default Filters
- Status
- Organization
- Role
- Invitation State
- Last Active
Saved Views
- Pending Invitations
- Suspended Users
- Recently Active
- Locked Accounts
- Users Without Verified Membership
Row Actions
- View Details
- Change Role
- Revoke Sessions
- Suspend
- Restore
- Archive
High-risk actions should not appear as the first row action.
=== USER DETAIL PAGE ===
Header
- Name
- Masked identifier
- Status
- Primary organization
- Current role
- Primary approved action
Sections
- Overview
- Organizations and Workspaces
- Roles and Permissions
- Authentication and Sessions
- Workflow Activity
- Support History
- Audit History
Sensitive Fields
Mask or restrict according to verified permissions.
=== USER STATES ===
Invited
Invitation sent but not accepted.
Pending Verification
Identity or email verification incomplete.
Active
Normal account access.
Suspended
Access intentionally paused.
Locked
Access blocked because of a verified security or policy event.
Disabled
Account access disabled by verified operational policy.
Archived
Inactive record retained according to verified policy.
Deleted
Deletion state and retention behavior require verification.
Every state must define
- Meaning
- Allowed actions
- Restrictions
- Recovery
- Audit event
- Customer notification where required
=== ORGANIZATION LIST ===
Columns
- Organization
- Owner
- Status
- Plan
- Seats
- Usage
- Workspaces
- Created
- Actions
Filters
- Status
- Plan
- Region
- Usage level
- Payment state
- Owner
=== ORGANIZATION DETAIL ===
Sections
- Summary
- Members
- Workspaces
- Subscription
- Usage
- Billing
- Integrations
- Domains
- Security
- Support
- Activity
- Audit Logs
High-Risk Actions
- Transfer Ownership
- Suspend Organization
- Cancel Subscription
- Archive Organization
- Delete Organization
=== ROLES AND PERMISSIONS ===
Authorization Model
Use the verified backend model. Do not infer missing grants.
Role List Columns
- Role Name
- Purpose
- Scope
- Assigned Users
- Permission Count
- System or Custom
- Last Modified
Permission States
- Granted Directly
- Inherited
- Explicitly Restricted
- Not Applicable
Permission Groups
- User Management
- Organization Management
- Workflows
- Billing
- Support
- Security
- Audit
- Integrations
- Data Export
- Impersonation
=== PERMISSION MATRIX ===
Rows
Resources and operational capabilities
Columns
- View
- Create
- Edit
- Delete
- Export
- Approve
- Manage Access
- Manage Billing
- Impersonate
Requirements
- Explain inheritance.
- Use text and icons rather than color alone.
- Support keyboard navigation.
- Summarize high-risk access.
- Validate dangerous combinations.
- Show affected users before saving.
=== PERMISSION CHANGE REVIEW ===
Display
- Added permissions
- Removed permissions
- Inherited changes
- Affected users
- Affected scopes
- High-risk capabilities
- Effective time
- Audit event
Primary CTA
Apply Permission Changes
Secondary Action
Return to Editing
=== DATA TABLE SYSTEM ===
Capabilities
- Server-side sorting
- Server-side filtering
- Cursor pagination
- Saved views
- Column selection
- Comfortable and compact density
- Row selection
- Bulk actions
- Export
- Loading
- Empty
- Filtered empty
- Partial data
- Error
- Stale-data warning
Table URL State
Preserve filters, sorting, and pagination where appropriate.
Row Actions
Provide one frequent visible action and secondary actions inside an accessible menu.
=== FILTER SYSTEM ===
Filter Types
- Search
- Status
- Role
- Organization
- Date Range
- Plan
- Owner
- Risk
- Payment State
Rules
- Keep applied filters visible.
- Support Clear All.
- Preserve state.
- Use operational language.
- Avoid hidden defaults.
- Support saved views.
- Keep URLs shareable where authorized.
=== BULK-ACTION FLOW ===
Example
Assign Role to Selected Users
- Select users.
- Choose Assign Role.
- Validate eligibility.
- Identify excluded users.
- Preview affected users and scope.
- Confirm.
- Process asynchronously where required.
- Show completed, excluded, and failed records.
- Allow retry for failed records where safe.
- Record an audit event for every completed change.
Do not report complete success when some records fail.
=== AUDIT LOG ===
Columns
- Timestamp
- Actor
- Action
- Object
- Scope
- Result
- Source
- Details
Human-Readable Event
Maya Chen changed Jordan Lee's workspace role from Viewer to Editor.
Audit Detail
- Actor
- Action
- Object
- Scope
- Timestamp
- Result
- Source
- Before
- After
- Request or trace identifier
- Related events
Sensitive values and secrets must be masked.
=== BILLING ADMINISTRATION ===
Modules
- Subscriptions
- Invoices
- Payments
- Credits
- Usage
- Refunds
- Payment Failures
- Billing Contacts
Subscription Detail
- Organization
- Plan
- Status
- Billing period
- Renewal
- Seats
- Usage
- Credits
- Invoices
- Discounts
- Cancellation
- History
Every billing action must show
- Authorization
- Effective date
- Financial impact
- Customer impact
- Reason
- Confirmation
- Audit event
- Notification behavior
- Recovery when supported
=== REFUND FLOW ===
Required Information
- Invoice or payment
- Eligible amount
- Full or partial refund
- Reason
- Payment method
- Tax behavior
- Customer notification
- Audit event
All eligibility and financial behavior require verification.
=== LOG VIEWER ===
Supported Logs
- Workflow Runs
- API Requests
- Webhooks
- Authentication
- Integrations
- Jobs
- Billing Events
Columns
- Timestamp
- Severity
- Source
- Message
- Status
- Duration
- Request ID
- Related Entity
Capabilities
- Search
- Filters
- Time Range
- Structured View
- Raw View
- Line Wrapping
- Copy
- Authorized Export
- Secret Masking
- Retention Explanation
=== SUPPORT CONSOLE ===
Sections
- Customer Identity
- Organization
- Subscription
- Usage
- Recent Workflow Failures
- Support Tickets
- Internal Notes
- Roles and Permissions
- Integrations
- Audit History
- Impersonation
Internal Note Requirements
- Author
- Timestamp
- Visibility
- Category
- Edit History
- Retention
- Sensitivity
=== IMPERSONATION FLOW ===
Requirements
- Explicit permission
- Required reason
- Approved duration
- Visible impersonation banner
- Clear Exit Impersonation action
- Restricted sensitive actions
- Audit event
- Customer scope
- Approval where required
- Notification behavior where required
Banner Example
You are viewing NexusFlow as Jordan Lee in Acme Operations.
Do not allow silent impersonation.
=== DESTRUCTIVE-ACTION FRAMEWORK ===
Example
Suspend Organization
Confirmation Displays
- Organization name
- Active scope
- Affected workspaces
- Affected users
- Workflow impact
- Billing impact
- Reversibility
- Notification behavior
- Reason field
- Required permission
High-risk actions may require
- Typed organization name
- Secondary approval
- Delayed execution
- Audit event
- Recovery path
=== RESPONSIVE ARCHITECTURE ===
Desktop
Persistent sidebar, dense tables, split views, inspectors, and multi-panel dashboards.
Tablet
Collapsed navigation, priority columns, filter drawers, touch-safe actions, and simplified split views.
Mobile
- Urgent tasks only
- Dedicated entity detail screens
- Priority table columns
- Card or expandable record views
- Simplified bulk actions
- Visible tenant and environment context
- Protected destructive actions
- Mobile-keyboard-safe forms
Avoid shrinking desktop tables until they become unreadable.
=== ACCESSIBILITY REQUIREMENTS ===
- Semantic landmarks
- Logical headings
- Accessible navigation
- Skip links
- Keyboard navigation
- Visible focus
- Proper table headers
- Table captions
- Sort-state announcements
- Selection-state communication
- Accessible filter labels
- Persistent form labels
- Error relationships
- Modal and drawer focus management
- Non-color status indicators
- Status announcements
- Touch-target review
- Zoom and reflow
- Responsive reading order
- Reduced-motion support
- Accessible log formatting
- Permission explanations
- Destructive-action clarity
=== ANALYTICS EVENT MAP ===
Event
global_search_used
Properties
- entity_type
- result_count_bucket
- operator_role
- scope_type
Event
filter_applied
Properties
- module
- filter_id
- operator_role
Event
bulk_action_started
Properties
- action_type
- selected_count_bucket
- module
- scope_type
Event
bulk_action_completed
Properties
- action_type
- success_count_bucket
- failure_count_bucket
- excluded_count_bucket
Event
permission_changed
Properties
- role_id
- scope_type
- added_count
- removed_count
- high_risk_change
Event
impersonation_started
Properties
- operator_role
- target_scope_type
- approved_duration_bucket
Privacy
Do not send names, emails, payment details, secrets, tokens, raw logs, support-note content, or sensitive audit metadata into analytics.
=== ACCEPTANCE CRITERIA ===
Tenant Isolation
Given an operator has access only to Organization A,
when global search is used,
then results from Organization B are not returned or disclosed.
Bulk Role Change
Given twenty users are selected,
when three users are ineligible,
then the interface identifies the exclusions, previews the eligible impact, processes valid records, and reports partial results accurately.
Permission Change
Given an administrator adds a high-risk permission,
when saving,
then the interface shows affected users, scope, inherited changes, operational risk, and the resulting audit event.
Impersonation
Given an authorized support operator begins impersonation,
when the customer session opens,
then a visible banner shows the target user and organization, restricted actions remain unavailable, and Exit Impersonation remains accessible.
Destructive Action
Given an administrator suspends an organization,
when confirmation appears,
then the affected organization, workspaces, users, workflows, billing implications, reversibility, and notification behavior are visible.
=== REGRESSION TEST MATRIX ===
Roles
- Super Administrator
- Support Agent
- Billing Administrator
- Organization Administrator
- Read-Only Auditor
- Unauthorized User
Scopes
- Platform
- Organization
- Workspace
- Project
- Production
- Test
Entity States
- Active
- Pending
- Suspended
- Locked
- Archived
- Deleted
Table States
- Loading
- Empty
- Filtered Empty
- Error
- Partial Data
- Stale Data
- Large Result Set
- Exporting
Bulk Actions
- All Eligible
- Partial Eligibility
- Partial Failure
- Complete Failure
- Asynchronous Completion
Accessibility
- Keyboard
- Screen Reader
- Zoom
- Reflow
- Reduced Motion
- High Contrast Where Supported
Responsive
- Wide Desktop
- Desktop
- Laptop
- Tablet
- Mobile
=== QA CHECKLIST ===
Scope
- Verify tenant context.
- Verify workspace context.
- Verify environment context.
- Verify cross-tenant restrictions.
Authorization
- Verify every role.
- Verify every permission.
- Verify inheritance.
- Verify server enforcement.
- Verify denied states.
Tables
- Verify columns.
- Verify sorting.
- Verify filters.
- Verify pagination.
- Verify saved views.
- Verify selection.
- Verify bulk actions.
- Verify partial failure.
Details
- Verify entity status.
- Verify relationships.
- Verify actions.
- Verify activity.
- Verify audit history.
Billing
- Verify plans.
- Verify invoices.
- Verify payments.
- Verify credits.
- Verify refunds.
- Verify effective dates.
- Verify financial impact.
Audit
- Verify actor.
- Verify action.
- Verify object.
- Verify scope.
- Verify timestamp.
- Verify result.
- Verify before and after values.
- Verify masking.
Accessibility
- Verify landmarks.
- Verify headings.
- Verify keyboard navigation.
- Verify focus.
- Verify tables.
- Verify modals.
- Verify drawers.
- Verify responsive order.
=== RISKS ===
Risk
Cross-tenant data exposure
Mitigation
Enforce tenant authorization on the server, scope all queries, and test search, lists, exports, and direct URLs.
Risk
Hidden permission inheritance
Mitigation
Show direct and inherited permissions separately and preview downstream changes.
Risk
Unsafe impersonation
Mitigation
Require permission, reason, time limit, visible banner, restricted actions, clear exit, and audit logging.
Risk
Bulk-action partial failure
Mitigation
Validate eligibility, process safely, report per-record outcomes, and support retry where appropriate.
Risk
Billing mistake
Mitigation
Verify financial rules, effective dates, proration, authorization, confirmation, and audit events.
Risk
Sensitive logs are exposed
Mitigation
Mask secrets, restrict access, define retention, and review exports.
=== KNOWN LIMITATIONS ===
- Authorization rules require backend verification.
- Tenant isolation requires security testing.
- Billing behavior requires financial and functional validation.
- Audit integrity depends on the underlying implementation.
- Impersonation requires security and privacy review.
- Accessibility requires implementation testing.
- Responsive behavior requires device and browser testing.
- Analytics must be implemented and validated.
- The architecture does not guarantee security, compliance, operational efficiency, or lower support costs.
=== FINAL INTEGRITY NOTE ===
All roles, permissions, billing rules, account states, data-retention behavior, deletion behavior, audit semantics, security claims, compliance claims, support privileges, impersonation rules, and operational consequences must be verified before implementation or publication.
The admin architecture is designed to improve operational clarity, consistency, safety, and implementation readiness. It does not guarantee security, compliance, productivity, error reduction, support efficiency, or lower operational cost.
admin-panel-ui-architect.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
Admin Panel UI Architect helps developers, SaaS founders, backend teams, B2B startups, internal-tools builders, operations teams, product designers, and frontend developers create professional administration interfaces that are clear, scalable, auditable, permission-aware, and safe to operate.
The skill transforms backend capabilities, entity models, database structures, role definitions, operational workflows, billing systems, support processes, or disorganized legacy dashboards into a coherent admin-product architecture.
It begins by defining the authorized operators, their responsibilities, task frequency, access scope, operational risk, data sensitivity, and escalation paths. Instead of mirroring raw backend modules, it organizes the interface around real administrative jobs.
The agent can design admin panels for SaaS applications, marketplaces, e-commerce platforms, B2B software, AI tools, content platforms, booking systems, education products, healthcare operations tools, finance workflows, subscription businesses, customer-support systems, logistics platforms, inventory systems, creator platforms, developer products, moderation consoles, partner portals, reseller portals, internal tools, and enterprise control systems.
It creates task-oriented navigation, tenant and workspace switching, environment indicators, global search, operational dashboards, entity inventories, list pages, detail pages, create and edit flows, settings, logs, reports, notifications, queues, and action workflows.
For multi-tenant products, the skill keeps the active organization, workspace, project, region, environment, and operator role visible. It defines safe tenant switching, scope-aware actions, cross-tenant protections, persistent navigation context, and additional confirmation for high-impact operations.
User-management systems can include invitations, verification states, role assignment, organization membership, session management, authentication methods, account suspension, account restoration, access revocation, archival, deletion, export, and support history.
Organization and workspace management can include owners, members, subscriptions, seats, usage, integrations, domains, security settings, billing contacts, plan limits, account states, support history, activity, and audit records.
The skill creates role and permission architectures for role-based, attribute-based, ownership-based, resource-level, environment-based, or hybrid authorization models. It defines role lists, permission groups, inheritance, direct grants, explicit restrictions, member assignments, high-risk capabilities, change reviews, and human-readable permission matrices.
Permission changes can include previews of added access, removed access, inherited changes, affected users, affected resources, operational impact, elevated-risk permissions, confirmation requirements, and audit events.
The agent creates complete data-table systems for scanning, comparison, filtering, selection, sorting, pagination, density control, column management, export, row actions, and bulk actions. It avoids displaying every database field and instead prioritizes the information required for operational decisions.
Table states can include loading, empty, filtered empty, stale data, partial data, permission restricted, offline, exporting, updating, and failure.
Filter systems can include search, status, role, organization, date range, plan, owner, region, risk, source, amount, tags, and custom operational filters. Filters can persist in the URL, remain active during navigation, appear as removable chips, and become saved views.
Saved views can support private, team, organization, and system visibility with preserved filters, sorting, columns, density, ownership, sharing, and default-view behavior.
Bulk operations are designed with eligibility checks, excluded-record handling, impact previews, confirmation, progress, partial success, failure recovery, retry, undo where possible, and per-entity audit records.
The skill creates structured detail pages with summary information, entity status, identifiers, relationships, permissions, billing, integrations, activity, notes, files, usage, support records, risk indicators, metadata, and audit history.
It designs dedicated create and edit flows, drawers, modals, inline editing, or full pages according to task complexity, operational risk, mobile behavior, and audit requirements. Complex or dangerous workflows are not forced into small modal windows.
Settings architecture can cover organization configuration, users, roles, authentication, security, billing, plans, integrations, APIs, notifications, branding, localization, data, developer settings, advanced settings, and audit configuration.
The skill creates audit-log systems that clearly communicate who performed an action, what changed, which object was affected, the operational scope, time, source, outcome, previous values, new values, and related identifiers.
It distinguishes user-friendly activity history from security-grade audit records and avoids exposing raw technical event names without interpretation.
Billing administration can include subscriptions, plans, invoices, payments, credits, usage, discounts, refunds, taxes, payment methods, failed payments, dunning, billing contacts, enterprise contracts, cancellations, reactivations, and change history.
All billing-related actions are designed with verified authorization, effective dates, financial impact, customer impact, reason capture, confirmation, audit records, notifications, and recovery where supported.
The agent can create reports for users, revenue, usage, churn, support volume, moderation, security, operations, compliance, and system performance. Reports include metric definitions, time zones, filters, freshness, missing-data behavior, exports, and limitations.
Log viewers can support application logs, API logs, authentication events, integrations, webhooks, jobs, email, billing, and system events. They include search, severity, source, status, duration, identifiers, structured and raw views, secret masking, wrapping, retention explanations, and permission controls.
Moderation and review consoles can include flagged content, reported users, transaction review, identity review, policy review, support escalation, and approval queues. Queue items can include reason, priority, risk, age, evidence, owner, actions, notes, escalation, and appeal requirements.
Support consoles can connect customer identity, organization, plan, usage, recent errors, tickets, activity, permissions, billing, integrations, internal notes, and controlled impersonation.
Impersonation workflows are treated as high risk. The skill defines explicit authorization, required reason, duration, visible impersonation banners, exit controls, restricted actions, audit logging, sensitive-data protections, approvals, and customer notifications where required.
Destructive actions such as deleting users, removing organizations, cancelling subscriptions, refunding payments, revoking access, disconnecting integrations, rotating credentials, or purging data are designed with risk-based confirmation levels, consequence explanations, dependency checks, typed confirmation where justified, reason capture, approval, delay, audit, and recovery.
The skill also creates permission-denied states, first-use empty states, filtered-empty states, error states, background-job states, partial-result states, responsive specifications, mobile record views, accessible table alternatives, analytics event maps, operational metrics, audit reports, findings, implementation sequences, acceptance criteria, regression tests, QA checklists, and risk assessments.
It prioritizes operational clarity, verified authorization, persistent scope, safe actions, auditability, human-readable system state, recoverable workflows, privacy, accessibility, responsive behavior, and implementation readiness.
The core commercial promise is: design admin panels that help authorized operators find the right information, understand system state, take safe actions, manage permissions, review history, and complete operational work with less confusion and lower risk.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 2 months ago
- Passed all security checks, Safe to install