Labsco
exploreomni logo

omni-ai-optimizer

★ 6

by exploreomni · part of exploreomni/omni-cursor-plugin

Optimize your Omni Analytics model for Blobby, Omni's AI assistant — configure ai_context, ai_fields, sample_queries, and create AI-specific topic extensions.…

🔥🔥🔥✓ VerifiedFreeQuick setup
#analytics#payments#git
🧩 One of 7 skills in the exploreomni/omni-cursor-plugin package — works on its own, and pairs well with its siblings.

This is the playbook your agent receives when the skill activates — you don't need to read it to use the skill, but it's here to audit before installing.

by exploreomni

Optimize your Omni Analytics model for Blobby, Omni's AI assistant — configure ai_context, ai_fields, sample_queries, and create AI-specific topic extensions.… npx skills add https://github.com/exploreomni/omni-cursor-plugin --skill omni-ai-optimizer Download ZIPGitHub6

Omni AI Optimizer

Optimize your Omni semantic model so Blobby (Omni's AI assistant) returns accurate, contextual answers.

Tip: Use omni-model-explorer to inspect current AI context before making changes.

API Discovery

When unsure whether an endpoint or parameter exists, fetch the OpenAPI spec:

curl -L "$OMNI_BASE_URL/openapi.json" \
 -H "Authorization: Bearer $OMNI_API_KEY"

Use this to verify endpoints, available parameters, and request/response schemas before making calls.

How Blobby Works

Blobby generates queries by examining:

  • Topic structure — which views and fields are joined

  • Field labels and descriptions — how fields are named

  • synonyms — alternative names for fields

  • ai_context — explicit instructions you write

  • ai_fields — which fields are visible to AI

  • sample_queries — example questions with correct queries

  • Hidden fields — hidden: true fields are excluded

Impact order: ai_context > ai_fields > sample_queries > synonyms > field descriptions.

Writing ai_context

Add via the YAML API:

curl -L -X POST "$OMNI_BASE_URL/api/v1/models/{modelId}/yaml" \
 -H "Authorization: Bearer $OMNI_API_KEY" \
 -H "Content-Type: application/json" \
 -d '{
 "fileName": "order_transactions.topic",
 "yaml": "base_view: order_items\nlabel: Order Transactions\nai_context: |\n Map \"revenue\" → total_revenue. Map \"orders\" → count.\n Map \"customers\" → unique_users.\n Status values: complete, pending, cancelled, returned.\n Only complete orders for revenue unless specified otherwise.",
 "mode": "extension",
 "commitMessage": "Add AI context to order transactions topic"
 }'

What Makes Good ai_context

Terminology mapping — map business language to field names:

ai_context: |
 "revenue" or "sales" → order_items.total_revenue
 "orders" → order_items.count
 "customers" → users.count or order_items.unique_users
 "AOV" → order_items.average_order_value

Data nuances — explain what isn't obvious from field names:

ai_context: |
 Each row is a line item, not an order. One order has multiple line items.
 total_revenue already excludes returns and cancellations.
 Dates are in UTC.

Behavioral guidance — direct common patterns:

ai_context: |
 For trends, default to weekly granularity, sort ascending.
 For "top N", sort descending and limit to 10.

Persona prompting — set the analytical perspective:

ai_context: |
 You are the head of finance analyzing customer payment data.
 Default to monetary values in USD with 2 decimal places.

Curating Fields with ai_fields

Reduce noise for large models:

ai_fields:
 - all_views.*
 - -tag:internal
 - -distribution_centers.*

# Or explicit list
ai_fields:
 - order_items.created_at
 - order_items.total_revenue
 - order_items.count
 - users.name
 - users.state
 - products.category

Same operators as topic fields: wildcard (*), negation (-), tags (tag:).

Adding sample_queries

Teach Blobby by example. Build the correct query in a workbook, retrieve its structure, then add to the topic YAML:

sample_queries:
 - prompt: "What month has the highest sales?"
 ai_context: "Use total_revenue grouped by month, sorted descending, limit 1"
 query:
 fields:
 order_items.created_at[month]: created_month
 order_items.total_revenue: total_revenue
 base_view: order_items
 sorts:
 - field: order_items.total_revenue
 desc: true
 limit: 1
 topic: order_transactions

Focus on questions users actually ask — check Analytics > AI usage in Omni.

AI-Specific Topic Extensions

Create a curated topic variant for Blobby using extends:

# ai_order_transactions.topic
extends: [order_items]
label: AI - Order Transactions

fields:
 - order_items.created_at
 - order_items.status
 - order_items.total_revenue
 - order_items.count
 - users.name
 - users.state
 - products.category

ai_context: |
 Curated view of order data for AI analysis.
 [detailed context here]

sample_queries:
 - prompt: "Top selling categories last month?"
 query:
 fields:
 products.category: category
 order_items.total_revenue: revenue
 base_view: order_items
 filters:
 order_items.created_at: "last month"
 sorts:
 - field: order_items.total_revenue
 desc: true
 limit: 10
 topic: ai_order_transactions

Improving Field Descriptions

dimensions:
 status:
 label: Order Status
 description: >
 Current fulfillment status. Values: complete, pending, cancelled, returned.
 Use 'complete' for revenue calculations.

Good descriptions help both Blobby and human analysts.

Adding synonyms

Map alternative names, abbreviations, and domain-specific terminology so Blobby matches user queries to the correct field. Works on both dimensions and measures.

dimensions:
 customer_name:
 synonyms: [client, account, buyer, purchaser]
 order_date:
 synonyms: [purchase date, transaction date, order timestamp]

measures:
 total_revenue:
 synonyms: [sales, income, earnings, gross revenue, top line]
 average_order_value:
 synonyms: [AOV, avg order, basket size]

Synonyms vs ai_context: Use synonyms for field-level name mapping. Use ai_context for topic-level behavioral guidance, data nuances, and multi-field relationships.

Optimization Checklist

  • Inspect current state with omni-model-explorer

  • Check AI usage dashboard for real user questions

  • Write ai_context mapping business terms to fields

  • Add synonyms to key dimensions and measures

  • Curate ai_fields to remove noise

  • Add sample_queries for top 3-5 questions

  • Improve field description values

  • Consider extends for AI-specific topic variants

  • Test iteratively — ask Blobby and refine

Docs Reference

Related Skills

  • omni-model-explorer — inspect existing AI context

  • omni-model-builder — modify views and topics

  • omni-query — test queries to verify Blobby's output