Works with the AI Agents you already use
cmdb data quality and relationship auditor
Broken CMDB data leads to failed changes, service outages, and wasted licensing costs.
$2.99
cmdb data quality and relationship auditor
Example session with this skill installed
Audit the attached JSON export of 500 Linux servers and their 'Depends On' relationships. Use analysis date 2023-11-01. Scope approved by Director of IT Ops. Check for duplicates and stale CIs.
- Read your context and instructions
- Compiled the cmdb data quality
The audit of 500 Linux server CIs and their relationships is complete. I have identified 12 duplicate candidates and 4 dependency cycles.
CMDB Data Quality Audit — IT Operations, 2023-11-01
1. Scope, sources and assumptions
- E1 (CIs): 500 supplied, 500 in scope.
- E2 (Relationships): 840 supplied, 840 evaluated.
- Extract Date: 2023-10-31.
- Assumptions: CIs not verified in 90 days are marked as stale.
2. Data quality dimensions
| Dimension | Numerator | Denominator | Rate |
|---|---|---|---|
| Completeness | 480 | 500 | 0.9600 |
| Accuracy (Duplicates) | 488 | 500 | 0.9760 |
| Consistency (Relationships) | 812 | 840 | 0.9666 |
| Integrity (Ownership) | 495 | 500 | 0.9900 |
Data Quality Index: 0.9732
3. Findings requiring attention
| Id | Severity | Code | CIs | Relationships | Message |
|---|---|---|---|---|---|
| F001 | High | CYCLE_DET | 8 | 8 | Dependency cycle detected in App-Cluster-01 |
| F002 | Med | STALE_REC | 15 | 0 | Records not verified in > 90 days |
| F003 | Low | MISSING_ATTR | 20 | 0 | IP Address missing on Linux class |
4. Duplicate candidates and dependency issues
- Management decision required: CI-992 and CI-1044 share serial number 'SN-99102'. Potential duplicate.
- Management decision required: 4 circular dependencies found in the web tier.
5. Ownership summary
- 48 groups identified. 3 groups suppressed (size < 5).
- 5 CIs assigned to inactive owner 'admin_user_04'.
6. Proposed actions (drafts)
| Action | Finding | Targets | Status |
|---|---|---|---|
| MERGE | F002 | CI-992, CI-1044 | DRAFT |
| UPDATE_OWNER | F005 | CI-102, CI-105 | DRAFT |
7. Warnings, limits and items for human review
- Information gap — organizational validation required: 20 CIs missing 'Environment' tag.
- This audit supports, and does not replace, professional judgment and management's decision.
Next steps
- Review the MERGE candidates in Section 6 against your primary discovery source.
- Resolve the dependency cycles in the App-Cluster-01 stack to fix impact analysis.
- Initiate a change record to update the 5 orphaned CIs with active owners.
Connects securely to your tools. The creator never sees your data.
About this skill
Broken CMDB data leads to failed changes, service outages, and wasted licensing costs. This skill audits your configuration management database to find duplicate records, orphaned relationships, and stale data using a deterministic, read-only approach.
What it does
- Audit completeness verifies that configuration items (CIs) contain all required attributes based on your class rules.
- Find duplicates identifies candidate records for merging based on hostname, serial numbers, and IP addresses.
- Validate ownership maps CIs to an owner directory to flag missing or inactive owners.
- Trace relationships detects dangling links, circular dependencies, and active CIs pointing to retired upstream services.
- Calculate metrics generates a formal data quality index with exact numerator and denominator rates.
How it works
- Source registration processes your CI exports, relationship records, and owner directories into a structured request.
- Scope validation checks for the required auditor approval and authorized CI classes before proceeding.
- Execution runs a deterministic Python audit script to perform graph analysis and duplicate matching.
- Draft reporting produces a structured audit report with a prioritized list of cleanup candidates and proposed actions.
Frameworks & tools
The skill utilizes a self-contained Python execution environment to process JSON-based CMDB extracts. It relies on graph-theory logic for dependency checks and follows a strict schema-driven input/output model.
Why this beats prompting it yourself
Manual prompts often hallucinate relationship counts or fail to detect complex circular dependencies in large datasets. This skill uses a logic-heavy script to ensure every rate, trace, and duplicate match is mathematically accurate and traceable to specific record IDs.
Use cases
- Identify duplicate server records across different discovery sources.
- Clean up orphaned "Runs On" relationships after a data center migration.
- Prepare for a CMDB certification project by finding CIs with missing mandatory fields.
- Audit service chains to find active applications depending on retired infrastructure.
Known limitations
Requires a structured export in JSON or CSV format. It cannot perform live discovery scans or write changes directly back to your CMDB.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Read-only; cannot perform live discovery or write to DB. Requires specific fields like status and class. Minimum group sizes apply to owner summaries to protect privacy.
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 3 days ago