sf-post-deploy-verification

Verifies deployed Salesforce metadata, Apex tests, data, and configuration against a live org.

2|Updated Sep 12, 2026
One-click install
npx skills add https://github.com/grzmol/vibe-force --skill sf-post-deploy-verification-grzmol
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sf-post-deploy-verification
Source: https://github.com/grzmol/vibe-force/tree/main/skills/sf-post-deploy-verification
Command: npx skills add https://github.com/grzmol/vibe-force --skill sf-post-deploy-verification-grzmol

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? A Salesforce deploy reporting "Succeeded" only proves metadata compiled, not that the feature works. This Skill closes that gap by running a structured post-deploy verification sequence against the real org: deploy result inspection, org Apex tests, anonymous-Apex smoke probes, data and configuration queries, limits checks, and a rollback-versus-forward-fix decision when verification fails. ## Core Features & Use Cases - Deploy and test verification: Reads sf project deploy report/resume results, runs org Apex tests with coverage gates, and interprets component and test failure fields. - Smoke probes and org health: Executes anonymous-Apex probes (schema, permissions, integration, config, write-rollback) with VF_PROBE markers, plus SOQL/Tooling API queries for flows, scheduled jobs, permission sets, custom metadata, limits, and AsyncApexJob failures. - Failure triage and reporting: Provides a rollback vs forward-fix decision tree and writes a machine-readable report at .vibeforce/reports/smoke-<ISO>.json. - Use Case: After a quick deploy of an Order management feature to a sandbox, run the wave-4 sequence to confirm the deploy succeeded, tests pass coverage gates, the smoke probes return VF_PROBE_OK, permission set assignments landed, and no new async job failures exist before marking the story done. ## Quick Start Ask the assistant to verify the last deployment to the vf-int org by running the post-deploy verification sequence and producing the smoke report.

Frequently Asked Questions about sf-post-deploy-verification

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify a Salesforce deployment actually worked after deploy?

Run sf project deploy report with the job id to confirm status Succeeded and zero component errors, then execute org Apex tests with sf apex run test and anonymous-Apex smoke probes via sf apex run --file. Finish with SOQL queries for configuration and a limits check.

How to run anonymous Apex smoke tests against a Salesforce org?

Write a probe script that emits VF_PROBE marker lines at LoggingLevel.ERROR and throws on failure, then run it with sf apex run --file <script> --target-org <alias> --json. Check result.compiled and result.success in the JSON output to determine pass or fail.

What is the difference between sf project deploy report and deploy resume?

sf project deploy report checks deploy status without updating source tracking, while sf project deploy resume resumes watching a timed-out or async deploy and does update source tracking. Job ids remain valid for 10 days from the start of the deploy.

Can smoke probes run safely against a production Salesforce org?

Yes, if probes are read-only and guard any DML behind an Organization.IsSandbox check. Probes that must write data wrap changes in Database.setSavepoint and Database.rollback, and callout probes target sandbox endpoints only.

When should I roll back versus forward fix a failed Salesforce deployment?

Roll back when there is data corruption, security exposure, or an undeactivatable broken feature, by deploying the previous known-good source. Forward fix when the deploy failed without applying, the breakage is contained, or a fix fits within one deploy cycle.