mcpField
S
sf-datacloud-segment

sf-datacloud-segment

github.com/jaganproData & databases424975 installsUpdated 26w ago
Repository

About sf-datacloud-segment

sf-datacloud-segment is a skill for Salesforce Data Cloud audience and insight tasks. It covers segment and calculated-insight creation, publishing, execution, verification, and troubleshooting, with readiness checks before changes.

Key features

  1. 01Create and inspect audience segments
  2. 02Create and run calculated insights
  3. 03Publish segments and check member counts
  4. 04Check org readiness
  5. 05Verify results with SQL

Use cases

  1. 01Create reusable audience segments
  2. 02Publish a segment and confirm its member count
  3. 03Run a calculated insight and verify aggregates
  4. 04Troubleshoot segment SQL

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/jaganpro/sf-skills --skill sf-datacloud-segment

After installing, type/sf-datacloud-segmentin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

CREDITS.md
LICENSE
README.md

Details

From the project README

sf-datacloud-segment: Data Cloud Segment Phase

Use this skill when the user needs audience and insight work: segments, calculated insights, publish workflows, member counts, or troubleshooting Data Cloud segment SQL.

When This Skill Owns the Task

Use sf-datacloud-segment when the work involves:

  • sf data360 segment *
  • sf data360 calculated-insight *
  • segment publish workflows
  • member counts and segment troubleshooting
  • calculated insight execution and verification

Delegate elsewhere when the user is:

Required Context to Gather First

Ask for or infer:

  • target org alias
  • unified DMO or base entity name
  • whether the user wants create, publish, inspect, or troubleshoot
  • whether the asset is a segment or calculated insight
  • expected success metric: member count, aggregate value, or publish status

Core Operating Rules

  • Treat Data Cloud segment SQL as distinct from CRM SOQL.
  • Run the shared readiness classifier before mutating audience assets: node ~/.claude/skills/sf-datacloud/scripts/diagnose-org.mjs -o <org> --phase segment --json.
  • Prefer reusable JSON definitions for repeatable segment and CI creation.
  • Use --api-version 64.0 when segment creation behavior is unstable on newer defaults.
  • Verify with counts or SQL after publish/run steps instead of assuming success.
  • Use SQL joins rather than segment members when readable member details are needed.

Recommended Workflow

1. Classify readiness for segment work
node ~/.claude/skills/sf-datacloud/scripts/diagnose-org.mjs -o <org> --phase segment --json
2. Inspect current state
sf data360 segment list -o <org> 2>/dev/null
sf data360 calculated-insight list -o <org> 2>/dev/null
3. Create with reusable JSON definitions
sf data360 segment create -o <org> -f segment.json --api-version 64.0 2>/dev/null
sf data360 calculated-insight create -o <org> -f ci.json 2>/dev/null
4. Publish or run explicitly
sf data360 segment publish -o <org> --name My_Segment 2>/dev/null
sf data360 calculated-insight run -o <org> --name Lifetime_Value 2>/dev/null
5. Verify with counts or SQL
sf data360 segment count -o <org> --name My_Segment 2>/dev/null
sf data360 query sql -o <org> --sql 'SELECT COUNT(*) FROM "UnifiedssotIndividualMain__dlm"' 2>/dev/null

High-Signal Gotchas

  • Segment creation can require --api-version 64.0.
  • segment members returns opaque IDs; use SQL joins when human-readable member details are needed.
  • Segment SQL is not SOQL.
  • Calculated insight assets and segment SQL have different limitations.
  • Publish/run steps may kick off asynchronous work even when the command returns quickly.
  • An empty segment or calculated-insight list usually means the module is reachable but unconfigured, not unavailable.

Output Format

Segment task: <segment / calculated-insight>
Action: <create / publish / inspect / troubleshoot>
Target org: <alias>
Artifacts: <definition files / commands>
Verification: <member count / query result / publish state>
Next step: <act / retrieve / follow-up>

References

Comments

Markdown supported