- Home
- Skills
- Data & Databases
- Salesforce Data360 Data Quality
Works with the AI Agents you already use
Salesforce Data360 Data Quality
Audit Salesforce Data Cloud (Data 360) ingestion and DMO integrity with evidence-based reconciliation and drift detection.
Free
Salesforce Data360 Data Quality
Example session with this skill installed
Audit the 'Sales_Order_Stream' in the 'Production' data space. Reconcile the S3 source counts against the 'SalesOrder_DMO' for the latest run.
- Read your context and instructions
- Compiled the salesforce data360 data
I have completed the data quality audit for the Sales_Order_Stream ingestion pipeline. The audit identified a 2.4% record loss during the transformation layer from DLO to DMO.
Data Quality Audit Report
| Check ID | Asset Name | Grain | Threshold | Observed | Result | Method / Query |
|---|---|---|---|---|---|---|
| DQ-001 | Sales_Order_S3 | Source | 100% | 50,000 | PASS | SELECT COUNT(*) FROM S3_Manifest |
| DQ-002 | Sales_Order_DLO | Ingestion | >99.9% | 49,985 | PASS | 15 records in Problem DLO (Error: 403) |
| DQ-003 | SalesOrder_DMO | Modeled | ±0% | 48,800 | FAIL | SELECT COUNT(*) FROM SalesOrder_DMO |
| DQ-004 | SalesOrder_DMO | PK Uniqueness | 0 Duplicates | 0 | PASS | GROUP BY PrimaryKey HAVING COUNT > 1 |
| DQ-005 | SalesOrder_DMO | Freshness | < 4h | 6.2h | FAIL | MAX(DataSourceObjectRefreshDate) |
Findings Summary
Source to DLO: Minimal loss (15 records). Problem DLO indicates Type Mismatch on the TotalAmount field for these records.
DLO to DMO: Significant variance of 1,185 records. Investigation of the mapping logic shows a JOIN filter on Account_ID is excluding records where the parent Account has not yet been ingested into the Individual DMO.
- Freshness: The latest refresh exceeded the 4-hour threshold due to a delay in the upstream S3 file arrival.
Next steps
- Update the
TotalAmountfield mapping to handle scientific notation found in source files. - Review the
Account_IDrelationship requirements; consider changing the join to aLEFT JOINor adjusting the ingestion sequence. - Trigger a manual refresh of the
SalesOrder_DMOonce the parent Account records are processed.
Connects securely to your tools. The creator never sees your data.
About this skill
Data quality issues in Salesforce Data Cloud often remain hidden until downstream consumers report broken dashboards or failed identity resolution. This skill provides a systematic framework for inspecting data integrity across the ingestion pipeline from Source to DMO.
What it does
- Schema Validation identifies drift in file and stream schemas, checking keys, event times, and refresh modes against approved contracts.
- Error Analysis inspects Problem Record DLOs to group rejects by error code and field without exposing sensitive personal data.
- Integrity Auditing detects null or duplicate primary keys, type/length non-conformance, and unexpected volume shifts.
- Reconciliation traces record counts through filters, transforms, and one-to-many mappings to isolate actual data loss.
- Relationship Mapping validates foreign keys after transformation to report orphan or ambiguous matches.
How it works
- Initialize by reviewing the project's data contract and recent run history to establish thresholds.
- Execute Queries against Source, DLO, and DMO layers to observe actual record states and refresh status.
- Trace Failures to the earliest possible layer in the pipeline where evidence of the issue exists.
- Report findings using the standard quality contract format, including environment and asset identifiers.
Frameworks & tools
This skill is designed for Salesforce Data Cloud (Data 360), focusing on Data Lake Objects (DLO), Data Model Objects (DMO), and ingestion streams.
Why this beats prompting it yourself
Generic prompts often miss the nuances of Data Cloud's multi-layer architecture, such as accounting for upsert logic during reconciliation. This skill enforces strict evidence-based reporting, preventing the agent from assuming an "Active" status implies successful record processing.
Use cases
- Ingestion Debugging find why specific records are landing in the Problem Record DLO.
- Schema Drift Detection verify if source system changes broke the stream mapping.
- Integrity Cleanup identify orphan DMO records before running identity resolution.
- Audit Reporting generate a reconciliation report for a specific data space or run.
Known limitations
Does not perform automated repairs; findings must be routed to owners via governance protocols. Requires existing access to Data Cloud query interfaces.
Cannot perform automated remediation. Requires pre-existing query access to the Data Cloud environment. Evidence-based only; will not infer status for inaccessible assets.
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 any payment.
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
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 5 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts