> ## Documentation Index
> Fetch the complete documentation index at: https://docs.lumos.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Awesome Albus Prompts

> Browse general Albus prompts for exploring attribute coverage, org hierarchy, app usage and access request volume, with the expected output of each.

Overview

Albus is your AI agent for access data exploration and policy design — grounded in real assignment, usage, and access-request data. It drives least-privilege at scale with clear rationale for every recommendation, and it answers open-ended questions to reveal trends, anomalies, and risks across your environment.

**Supported Capabilities**

* **Policy Mining:** Discover common access patterns; draft least-privilege policies from real-world data.
* **Role Anomaly Detection:** Identify outlier or risky entitlements and unusual access patterns.
* **Data Exploration and Explainability:** Drill into usage signals, peer comparisons, and key differences.
* **Bulk Policy Creation and Edits:** Generate and tune access policies using natural language.

## Where to Start

* **General Prompts** (this page) — orient yourself, explore your environment, run quick lookups.
* [**Role Mining Prompts**](/for-admins/lumos-ai/role-mining-prompts) — build layered, threshold-based policies for joiner / mover / leaver automation.
* [**Security Findings Prompts**](/for-admins/lumos-ai/security-findings-prompts) — surface anomalies, risky access, drift, and over-provisioning.

The prompts below have customizable placeholders in `[brackets]`. Treat them as starting points — experiment, customize, and iterate to match your environment.

## General Prompts

| **Label** | **Prompt** | **Expected Output** | **Value** |
| - | - | - | - |
| Attribute coverage (active identities) | `Show me all user attributes and create a table with attribute type, number of values, and % of active users with values.` | A table listing each HR/custom attribute, its number of unique values, and coverage across active human identities. | Know which fields are reliable for policy segmentation; avoid sparse attributes. |
| Org hierarchy | `Return a tree-based diagram showing how key attributes relate to each other — that is, my organization hierarchy. Include user counts at each level.` | A text-based tree of org structure (e.g., business unit → department → team) with user counts. | Understand how deep your org goes before deciding policy granularity. |
| Reporting chain for a leader | `Give me every department that reports up to [manager name or email].` | A list of departments rolling up to that leader, with user counts per department. | Scope policies, reviews, or audits to a single leader's organization without guessing the boundary. |
| Org chart for a person | `Create a text-based tree showing the reporting structure starting with [individual email address].` | A reporting tree centered on that person, including direct and indirect reports. | See span of control when designing approvals and least-privilege decisions. |
| Department roster | `Who is in the [department name] department? Include manager and worker type for each person.` | A roster of people in the department with manager and worker-type context. | Validate department membership before scoping a policy to that group. |
| Manager lookup | `Who is [person's name]'s manager, and how many direct reports does that manager have?` | The manager's name, title, and direct-report count. | Quick sanity check when investigating a specific user's access or designing an approval flow. |
| Top apps by user count | `What are the top 10 apps by user count? Mark which look like core company-wide apps vs. team-specific.` | A ranked table of the most-assigned apps with a "core vs. niche" label. | Identify your must-have birthright candidates in one glance. |
| Top access requests | `Which apps have the highest number of access requests in the last quarter? Show the top 10 with app name and request count.` | A ranked table of the top 10 apps by request volume. | Focus policy and automation work where ticket pain is concentrated. |
| App deep-dive | `Show me the access patterns for [App Name] — who has it, how they break down by team and employee type, and how much it's actually being used.` | A summary of assignment, usage, and segmentation for the app. | Build context on a single app before drafting policies or anomaly reviews. |
| Look up an individual | `Who is the Head of [Function], and what apps and entitlements do they have?` | The person's profile and their access. | Quickly validate that key personas have the access you'd expect. |
| See what Albus has learned | `Show me my memories.` | A list of every fact, naming convention, and exclusion rule you've taught Albus. | Audit and clean up Albus's stored context so future recommendations reflect your current environment. |