- Home
- Skills
- Agents & Orchestration
- Salesforce Data360 Refresh Orchestration
Works with the AI Agents you already use
Salesforce Data360 Refresh Orchestration
Orchestrate Salesforce Data 360 refresh pipelines with dependency mapping and automated stage-gate verification.
Free
Salesforce Data360 Refresh Orchestration
Example session with this skill installed
Run the refresh pipeline for the 'Loyalty_Tier_Update' manifest. Check Data Streams first, then Calculated Insights. Use a 30-minute timeout per stage.
- Read your context and instructions
- Compiled the salesforce data360 refresh
I have initialized the refresh orchestration for the
Loyalty_Tier_Update manifest. The environment check passed, and I am proceeding with the stage-gate execution.
Pipeline Execution Summary
| Stage | Asset Name | Decision | Final State | Elapsed | Freshness Marker |
|---|---|---|---|---|---|
| Data Stream | AWS_S3_Loyalty_Events | Joined | SUCCESS | 04:12 | 2023-11-24T10:15:00Z |
| Data Stream | POS_Transaction_Logs | Started | SUCCESS | 08:45 | 2023-11-24T10:19:30Z |
| Calculated Insight | High_Value_Customer_Score | Started | SUCCESS | 12:20 | 2023-11-24T10:32:00Z |
| Segment | Premium_Tier_Members | Skipped | N/A | 00:00 | No Input Change |
| Activation | Email_Marketing_Push | Pending | WAITING | -- | -- |
Status Details:
- All 2 Data Streams completed successfully within the stage timeout.
- Calculated Insight
High_Value_Customer_Scorewas triggered as its inputPOS_Transaction_Logswas refreshed. - Segment
Premium_Tier_Memberswas skipped because match rules and input mappings remained unchanged per the scope logic. - The pipeline is currently waiting to trigger the final Activation stage.
Next steps
- Monitor the
Email_Marketing_Pushactivation in the Salesforce UI for external delivery confirmation. - Review the
POS_Transaction_Logsingestion logs if freshness markers do not align with source system offsets. - Verify the Calculated Insight output in the Data Explorer to confirm scoring accuracy.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing Salesforce Data 360 refreshes manually leads to fragmented data, circular dependencies, and wasted compute credits. This skill orchestrates post-change refresh pipelines in strict dependency order to ensure data integrity across your entire stack.
What it does
- Dependency mapping ensures jobs run in the correct sequence from Data Streams to Activations.
- Smart job joining prevents duplicate job submissions by polling existing pending runs.
- Terminal state verification requires explicit success timestamps rather than just "Active" status.
- Granular scope control skips unaffected assets to save processing time and resources.
- Automated stage gating halts downstream execution immediately upon upstream failure or timeout.
How it works
- Manifest creation where you define the specific assets and dependencies affected by your changes.
- Plan verification to audit the environment, stage order, and defined timeouts before execution.
- Stage execution following the strict hierarchy: Streams, Transforms, Identity, Insights, Segments, then Activations.
- Final reporting summarizing job trigger decisions, elapsed time, and freshness markers.
Frameworks & tools
Built for Salesforce Data Cloud (Data 360) environments using Node.js for orchestration logic. It utilizes the project scripts/run-refresh-pipeline.mjs for deterministic execution and monitoring.
Why this beats prompting it yourself
Generic prompts often fail to respect the strict hierarchy of Data Cloud objects or miss the nuance between an "Active" status and a truly finished refresh. This skill enforces a hard pipeline contract that handles polling, timeouts, and state verification without human intervention.
Use cases
- Orchestrating refreshes after updating Identity Resolution match rules.
- Triggering downstream segments and activations after a Calculated Insight definition change.
- Running a full pipeline refresh following new Data Stream ingestion.
- Safely joining and monitoring long-running jobs to prevent credit waste from redundant triggers.
Known limitations
Requires an authorized source publication or live model change before use. Does not support bypassing platform UI-only actions.
Requires existing authorized source publication. Cannot bypass UI-only platform actions. Stops on unknown states or manual handoffs.
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