/system-sync Skill

Sync systems and architecture data with external sources (CMDB, Confluence, ServiceNow, Jira).

When to Use This Skill

Use /system-sync when you need to:

•Sync system metadata from enterprise CMDB
•Sync architecture documentation from Confluence
•Sync ADRs from organization governance systems
•Synchronize with multiple data sources
•Create single source of truth for architecture data
•Validate vault data against authoritative sources
•Detect drift between local and remote definitions

Usage

code

/system-sync <source> [filter] [options]

Parameters

Parameter	Description	Required
`source`	Data source (cmdb, confluence, servicenow, jira, all)	Yes
`filter`	Filter by type, tag, or search term	Optional
`--mode`	Sync mode (fetch, validate, merge, publish)	Optional
`--direction`	Direction (pull, push, bidirectional)	Optional
`--dry-run`	Preview changes without applying	Optional

Supported Sources

1. CMDB (Configuration Management Database)

Typical source: ServiceNow CMDB, BMC CMDB, or custom CMDB

What syncs:

•Systems (name, type, owner, environment)
•System dependencies (connectsTo relationships)
•System metadata (IP addresses, hosts, ports)
•Deployment information (which account/region)
•Operational contacts (on-call, support)

Examples:

code

/system-sync cmdb --filter "type:Application"          # Sync all applications
/system-sync cmdb --filter "environment:production"    # Production systems only
/system-sync cmdb --dry-run                            # Preview changes

2. Confluence

Typical source: YourOrg Enterprise Confluence (wiki documentation)

What syncs:

•Architecture documentation (HLD, LLD, ADRs)
•System catalogs and inventories
•Integration documentation
•Governance policies and guardrails
•Strategic roadmaps
•Decision records

Examples:

code

/system-sync confluence --filter "space:ARCHITECTURE"  # All architecture docs
/system-sync confluence --filter "label:adr"           # All ADRs
/system-sync confluence --dry-run

3. ServiceNow

Typical source: YourOrg Service Management (change, incident, CMDB)

What syncs:

•Service mappings (services and supporting CIs)
•Change calendars (what's changing when)
•Incident history (what systems failed when)
•Change management data (deployments, cutover plans)
•SLA definitions

Examples:

code

/system-sync servicenow --filter "service:DataPlatform"        # DataPlatform service and dependents
/system-sync servicenow --mode validate                # Check for conflicts

4. Jira

Typical source: YourOrg Engineering Jira (projects, issues, roadmaps)

What syncs:

•Project metadata (status, team, roadmap)
•Epic/Feature definitions
•ADRs as Jira records
•Technology decisions from issues
•Release planning

Examples:

code

/system-sync jira --filter "project:MyDataIntegration"            # MyDataIntegration project info
/system-sync jira --filter "type:ADR"                  # ADRs in Jira

Sync Modes

1. Fetch (Pull from Source)

code

/system-sync cmdb --mode fetch

What happens:

•Queries source system
•Reads all matching records
•Creates/updates local notes
•Marks records with source: cmdb, lastSynced: <timestamp>
•Preserves local edits (non-authoritative fields)

Local fields preserved:

•Analysis and commentary
•Architecture context
•Decision rationale
•Local relationships

Remote fields updated:

•System name, type, owner
•Dependencies and relationships
•Metadata (IP addresses, hosts)
•Contact information

2. Validate (Check for Conflicts)

code

/system-sync cmdb --mode validate

What happens:

•Queries source system
•Compares with local notes
•
Reports discrepancies:
- •System exists locally but not in CMDB
- •System exists in CMDB but not locally
- •Metadata differs (owner, criticality, etc.)
- •Dependencies differ
•Generates validation report

Output example:

code

Validation Results: CMDB
────────────────────────────────
Total Systems: 25
✓ In Sync: 22
⚠ Out of Sync: 2
✗ Inconsistent: 1

Out of Sync:
- SAP S/4HANA
  • Local: criticality=critical
  • CMDB: criticality=high
  • Action: Update local or CMDB

Inconsistent:
- DataPlatform
  • Local: owner=John Smith
  • CMDB: owner=Data Platform Team
  • Action: Investigate

3. Merge (Bidirectional Reconciliation)

code

/system-sync cmdb --mode merge --dry-run

What happens:

•Compares local and remote
•
Merges data intelligently:
- •Takes more recent data
- •Preserves local analysis/context
- •Updates relationships
•Resolves conflicts interactively
•Creates merged records

Conflict resolution:

•If local and remote both changed: User chooses which to keep
•If only local changed: Keeps local
•If only remote changed: Takes remote
•If both unchanged: No action needed

4. Publish (Push to Source)

code

/system-sync confluence --mode publish

What happens:

•Reads local notes (System, Integration, Architecture)
•Converts to target format (Confluence pages, Jira issues, etc.)
•Creates/updates records in target system
•Tracks publication in publishedUrl field

Publication options:

•Create new pages/records (if not published before)
•Update existing (if already published)
•Archive deprecated records

Workflow

Phase 1: Configure Source

User specifies:

•
Source system
- •CMDB, Confluence, ServiceNow, or Jira
- •Which instance/environment
- •Authentication credentials (OAuth token, API key)
•
Sync filter
- •Which records to sync (type, tag, search)
- •Optionally: Which fields to sync
•
Sync mode
- •Fetch (pull), Validate (check), Merge (reconcile), Publish (push)
•
Dry-run?
- •Preview changes without applying

Phase 2: Execute Sync

The skill:

•
Connects to source
•
Queries records matching filter
•
For each record:
- •Find matching local note (by ID, name, or fuzzy match)
- •Apply sync mode (fetch/validate/merge/publish)
- •Track changes
•
Creates sync report showing:
- •Records processed
- •Records created
- •Records updated
- •Records with conflicts
- •Errors (if any)

Phase 3: Review and Confirm

If dry-run:

•Show preview of all changes
•User confirms or cancels
•If confirmed, apply changes for real

If not dry-run:

•Apply changes immediately
•Show summary report

Phase 4: Post-Sync Actions

Optionally:

•Create summary notes for new systems
•Generate diagrams for new architectures
•Trigger validation queries (e.g., find orphaned notes)
•Notify stakeholders of sync results

Examples

Example 1: Fetch CMDB Systems

code

/system-sync cmdb --filter "type:Application" --dry-run

Preview Output:

code

Sync Mode: Fetch (Pull from CMDB)
Filter: type="Application"
Dry-Run: Yes (preview only)

Proposed Changes:
────────────────
Create (New):
  ✓ ServiceNow (Service Management)
  ✓ Splunk (Log Management)

Update (Existing):
  ✓ SAP S/4HANA (update criticality: high → critical)
  ✓ DataPlatform (update owner, contacts)
  ✓ Snowflake (update deployment region)

Total: 5 records affected

User confirms → Changes applied:

•Creates 2 new System notes
•Updates 3 existing System notes
•All marked with source: cmdb, lastSynced: 2026-01-14T14:32:00Z

Example 2: Validate Confluence Architecture Docs

code

/system-sync confluence --filter "label:architecture" --mode validate

Output:

code

Validation Mode: Confluence Architecture Documentation
Filter: label="architecture"

Results:
────────
✓ In Sync: 8 ADRs
⚠ Out of Sync: 2 pages
✗ Missing: 1 page not in local vault

Out of Sync Details:
- "ADR - API Gateway Selection"
  • Confluence version: 5 (2026-01-10)
  • Local version: 4 (2026-01-08)
  • Recommendation: Fetch latest from Confluence

- "Architecture - Data Integration Platform"
  • Confluence: Minor edits only
  • Local: Significant additions (new scenarios)
  • Recommendation: Publish local version to Confluence

Missing from Vault:
- "Page - Authorization Strategy Review"
  • Last modified: 2025-12-15
  • Status: Active
  • Recommendation: Fetch into vault

Example 3: Publish Local ADRs to Confluence

code

/system-sync confluence --filter "type:Adr" --mode publish --direction push

Output:

code

Publication Mode: Local → Confluence
Filter: type="Adr"

Proposed Actions:
─────────────────
Create New Pages (Not Published):
  ✓ ADR - Snowflake as Analytics Warehouse
  ✓ ADR - Looker vs Tableau Selection
  ✓ ADR - Real-time Streaming Architecture

Update Existing Pages (Already Published):
  ✓ ADR - API Gateway Selection (version 5 → 6)
  ✓ ADR - SAP to AWS Connectivity (version 3 → 4)

Archive Deprecated (in Confluence, mark as deprecated):
  ✓ ADR - Tableau BI Platform (superseded by Looker)

Total: 10 pages affected

User confirms → Published to Confluence

Example 4: Bidirectional Merge with CMDB

code

/system-sync cmdb --mode merge

Intelligent merge:

•CMDB has updated system criticality → Update local
•Local has updated system owner in Confluence → Keep local
•Both have different SLA definitions → User resolves conflict

Result:

•Authoritative system metadata from CMDB
•Enriched with local analysis and context
•Conflict-free, single source of truth

Sync Report Structure

Each sync generates a detailed report:

Report Section 1: Summary

code

Sync Summary Report
═══════════════════════════════════════════════════════════
Source: CMDB (Configuration Management Database)
Mode: Fetch (Pull)
Filter: type="Application", environment="production"
Dry-Run: No (changes applied)
Timestamp: 2026-01-14T14:32:00Z

Records Processed:     25
Records Created:        3
Records Updated:        8
Records Unchanged:     14
Records with Errors:    0
─────────────────────────────────────────────────────────
Total Time: 3.2 seconds
Status: ✓ Success

Report Section 2: Created Records

code

New Records Created:
───────────────────
1. System - ServiceNow
   • Name: ServiceNow Service Management
   • Type: Platform
   • Owner: IT Service Management
   • Criticality: Medium
   • Status: Active

2. System - Splunk
   • Name: Splunk Log Management
   • Type: Platform
   • Owner: Operations & Engineering
   • Criticality: High
   • Status: Active

Report Section 3: Updated Records

code

Records Updated:
────────────────
1. System - SAP S/4HANA
   Changes:
   • criticality: high → critical
   • owner: "IT Applications" → "Finance IT"
   • dataClassification: internal → confidential

2. System - DataPlatform
   Changes:
   • contacts (support): [John, Sarah] → [John, Sarah, Mike]
   • status: active → active (no change)

Report Section 4: Errors/Conflicts

code

Sync Errors (0):
────────────────
None

Unresolved Conflicts (0):
─────────────────────────
None

Warnings (1):
─────────────
1. System "Cloud Foundry" found in CMDB but no matching local note
   Recommendation: Create new System note or add to exclusion filter

Sync Metadata

Records synced from external sources include:

yaml

# Sync Source Fields
source: cmdb | confluence | servicenow | jira
sourceId: "<unique-id-in-source>"
sourceUrl: "<url-to-source-record>"
sourceVersion: <version-number>
lastSynced: <ISO-timestamp>
syncMode: fetch | merge

# Authority
readOnly: false                    # Can local edits be synced back?
authoritative: <field-list>       # Which fields come from source

Advanced Options

Selective Field Sync

code

/system-sync cmdb --fields "owner,criticality,contacts" --dry-run

Only sync specific fields, preserve local edits to other fields.

Conflict Resolution Strategy

code

/system-sync cmdb --conflicts "ask|remote|local|newer|merge"

•ask - Prompt user for each conflict (default)
•remote - Always take source value
•local - Always keep local value
•newer - Take whichever was updated most recently
•merge - Intelligently merge (combine lists, keep both)

Schedule Automatic Syncs

code

/system-sync cmdb --schedule "daily at 02:00" --store

Configure recurring sync (stored in vault settings).

Exclude Records

code

/system-sync cmdb --exclude "status:deprecated,tag:archived"

Skip records matching filter.

Integration with Other Skills

The /system-sync skill works with:

•/system - Create/update System notes from synced data
•/integration - Update Integration notes with synced dependencies
•/diagram - Regenerate diagrams after sync
•/vault-maintenance - Detect orphaned notes after sync

Error Handling

Common sync errors and solutions:

Error	Cause	Solution
Authentication failed	Wrong credentials	Check API key/token in settings
No matching records	Filter too restrictive	Broaden filter or check source data
Cannot match records	Naming differs	Use source ID for matching, or manual map
Field not found	Field doesn't exist in source	Update field mapping config
Connection timeout	Network/firewall issue	Check network, retry, or use proxy

Sync Best Practices

•Always dry-run first - Preview changes before applying
•Sync during maintenance window - Minimize user impact
•Review sync reports - Check for unexpected changes
•Version control - Commit before/after syncs to git
•Monitor quality - Run validation sync periodically
•Document mappings - Keep field mappings documented
•Notify stakeholders - Let teams know when data synced

Sync Validation Queries

After syncing, run these queries to verify:

dataview

// Find systems with mismatched owners
TABLE owner, cmdbOwner
FROM ""
WHERE type = "System" AND owner != cmdbOwner

dataview

// Find orphaned local systems (not in CMDB)
TABLE source, sourceId
FROM ""
WHERE type = "System" AND source != "cmdb"

Next Steps

After syncing:

•Review sync report for any surprises
•Run validation queries to check consistency
•Update diagrams if topology changed (/diagram --refresh)
•Notify stakeholders of sync completion
•Schedule next sync cycle

Configuration

Sync sources configured in .claude/sync-config.yaml:

yaml

cmdb:
  url: "https://cmdb.yourorg.internal/api"
  auth: oauth2
  updateFrequency: daily

confluence:
  url: "https://confluence.yourorg.internal"
  space: ARCHITECTURE
  auth: oauth2

servicenow:
  instance: "yourorg"
  auth: oauth2

jira:
  url: "https://jira.yourorg.internal"
  projects: ["MyDataIntegration", "ModernizationProject", "DataPlatform"]
  auth: oauth2

Invoke with: /system-sync <source>

Example: /system-sync cmdb --filter "environment:production" --dry-run → Preview sync of production systems