mcpField
A
admin-settings-ia-review

admin-settings-ia-review

github.com/sylphxaiDesign & UI146 installsUpdated today
Repository

About admin-settings-ia-review

Admin Settings IA Review is a skill for structuring high-stakes administrative settings into reviewable design artifacts. It helps teams map settings, permissions, risky actions, user flows, and measurement plans.

Key features

  1. 01Inventory settings by scope and risk
  2. 02Group settings by user mental model
  3. 03Map permissions and role gates
  4. 04Specify confirmations and recovery paths
  5. 05Produce IA maps and state flows
  6. 06Plan instrumentation

Use cases

  1. 01Review an admin settings structure
  2. 02Clarify roles and access to controls
  3. 03Design safe flows for destructive actions

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/sylphxai/skills --skill admin-settings-ia-review

After installing, type/admin-settings-ia-reviewin Claude Code to confirm it loaded.

Bundled filesagents/openai.yamlreferences/admin-settings-ia-patterns.md

Details

From the project README

Admin Settings IA Review

Use this skill to convert a high-stakes product decision into a concrete, measurable, reviewable operating artifact.

Workflow

  1. Inventory settings by actor, scope, permission, risk, frequency, dependency, and audit need.
  2. Read references/admin-settings-ia-patterns.md.
  3. Group settings by user mental model: account, workspace, members, billing, security, integrations, data, notifications, support.
  4. Map dangerous actions, defaults, role gates, confirmation, recovery, and audit logs.
  5. Produce IA map, permission matrix, state flow, empty/error states, and instrumentation plan.

When not to use

  • Do not use for generic advice the base model already handles without this skill's specific artifact contract.

Guardrails

  • Do not bury destructive or security-critical controls behind ambiguous labels.
  • Do not let billing admins silently become security or data admins unless intended.
  • Avoid settings that depend on tribal knowledge, support-only changes, or hidden defaults.
  • Every dangerous action needs confirmation, consequence copy, permission check, and recovery path where possible.

Output format

Settings context:
Actors / scopes / risk:

IA map:
- <section>
  - <setting> -> <owner, scope, permission, default, audit>

Permission matrix:
| Setting | Viewer | Editor | Approver | Audit event | Recovery |
| --- | --- | --- | --- | --- | --- |

Risk review:
- <dangerous action> -> <confirmation/recovery/support>

Comments

Markdown supported