Works with the AI Agents you already use
role based access design
Manually designing RBAC roles from sprawling access spreadsheets is slow and prone to error.
$1.99
role based access design
Example session with this skill installed
Analyze the attached user_access.csv. Find over-privileged users and propose new roles for the Finance department using a min_users_for_role of 2.
- Read your context and instructions
- Compiled the role based access
The access analysis for the Finance department is complete. I have processed 45 user records and identified 3 potential role clusters.
RBAC Role Design Analysis — Finance Dept, 2023-10-27
1. Summary
- User Count: 45
- Entitlement Count: 112
- Average Entitlements/User: 8.4
- Mined Roles Count: 3
- Role Coverage: 82%
2. Mined Role Proposals
| Role ID | Entitlement Count | User Coverage |
|---|---|---|
| PROP_FIN_AP | 12 | 14 |
| PROP_FIN_AR | 10 | 12 |
| PROP_FIN_MGR | 22 | 4 |
3. Privilege Analysis
| User ID | Username | Type | Findings |
|---|---|---|---|
| U992 | jdoe | Over-privileged | 4 extra entitlements (sys_admin_shell, db_drop) |
| U881 | asmith | Under-privileged | Missing 2 core Finance entitlements |
4. Role Explosion Findings
- Oversized Roles: 1 role detected with > 15 entitlements (PROP_FIN_MGR).
- Single-User Roles: None detected.
- Duplicate Pairs: Role
EXISTING_FIN_CLERKandPROP_FIN_APshare 90% overlap.
5. Proposed Roles (require approval)
- Role: Finance Accounts Payable
- Entitlements:
fin_invoice_read,fin_invoice_write,erp_ap_access - Requires Human Approval: true
6. Warnings and Data Quality
WARNING: 5 users in the source data have no assigned job_role; they were excluded from privilege baseline analysis.
Next steps
- Review the
PROP_FIN_MGRrole to see if it should be split to reduce entitlement count. - Submit the
PROP_FIN_APproposal to the Finance Business Owner for formal sign-off. - Investigate user
jdoeto revoke thedb_droppermission.
Connects securely to your tools. The creator never sees your data.
About this skill
Manually designing RBAC roles from sprawling access spreadsheets is slow and prone to error. This skill automates role mining by analyzing user access patterns to identify logical role groupings and detect security risks like over-privileged accounts.
What it does
- Role Mining analyzes user access to propose logical role definitions based on entitlement overlaps.
- Privilege Auditing flags over-privileged users who have drifted from their job role baselines.
- Explosion Detection identifies redundant, oversized, or single-user roles that clutter IAM systems.
- Coverage Metrics computes exactly how much of the existing access landscape is covered by proposed roles.
- Business Approval Workflow prepares formal role proposals that designate the need for business-owner sign-off.
How it works
- Data Intake reads user lists, current entitlements, and job role baselines from your provided files.
- Analysis Execution runs a deterministic script to process the access matrix and identify clusters.
- Result Review checks the envelope for status codes like
completedorneeds_review. - Report Generation produces a structured breakdown of mined roles, privilege gaps, and explosion findings.
Frameworks & tools
The skill utilizes a Python-based analysis engine (scripts/run.py) and validates data against strict JSON schemas for access governance.
Why this beats prompting it yourself
Generic LLM prompts often hallucinate entitlements or miss subtle overlaps in large datasets. This skill uses a deterministic calculation script to ensure every mined role and privilege flag is backed by a full trace and exact data points.
Use cases
- Auditing user access to find accounts with excessive permissions.
- Consolidating fragmented access into a clean RBAC role structure.
- Reviewing role explosion during IAM migrations or post-merger integration.
- Preparing role design proposals for internal governance or audit reviews.
Known limitations
Does not perform segregation of duties (SoD) or toxic combination analysis. Not for identity reconciliation or joiner-mover-leaver workflows. Requires specific input fields like user_id and entitlements.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
No SoD/toxic combination analysis. Requires structured user/entitlement data. Does not perform identity reconciliation or JML workflow analysis.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today