Skip to main content

msh ai context

Generate AI-ready context pack for asset or project.

Usage​

msh ai context [--asset <asset_id>] [--json] [--include-tests] [--include-history]

Description​

Generates a comprehensive context pack containing:

  • Project information
  • Asset metadata
  • Lineage graph
  • Glossary terms
  • Schemas
  • Tests (optional)
  • Metrics and policies

Options​

  • --asset <asset_id>: Focus context pack on specific asset (includes upstream/downstream)
  • --json: Output context pack as JSON
  • --include-tests: Include test metadata in context pack
  • --include-history: Include recent run/deploy history

Examples​

Generate Project-Level Context Pack​

msh ai context

Generates context pack for entire project.

Generate Asset-Focused Context Pack​

msh ai context --asset revenue

Generates context pack focused on revenue asset, including upstream and downstream dependencies.

Include Tests and History​

msh ai context --asset revenue --include-tests --include-history

Includes test metadata and recent run/deploy history in context pack.

Output as JSON​

msh ai context --asset revenue --json

Example JSON Output:

{
  "project": {
    "id": "my-project",
    "name": "My Project",
    "warehouse": "postgres",
    "default_schema": "public"
  },
  "assets": [
    {
      "id": "revenue",
      "path": "assets/revenue.msh",
      "blocks": {
        "ingest": {...},
        "transform": {...}
      },
      "schema": {
        "columns": [
          {"name": "customer_id", "type": "integer"},
          {"name": "month", "type": "date"},
          {"name": "monthly_revenue", "type": "decimal"}
        ]
      }
    }
  ],
  "lineage": [
    {
      "from": "stg_orders",
      "to": "revenue",
      "type": "transform"
    }
  ],
  "glossary_terms": [
    {
      "id": "term.customer",
      "name": "Customer",
      "description": "A customer entity"
    }
  ],
  "schemas": {...},
  "tests": [...],
  "metrics": [...],
  "policies": [...]
}

Use Cases​

Preparing Context for AI Commands​

Context packs are automatically used by AI commands, but you can generate them manually:

# Generate context pack
msh ai context --asset revenue --json > context.json

# Use in custom AI workflows

Understanding Project Structure​

Generate context pack to understand project structure:

msh ai context --json | jq '.assets | length'

Analyzing Dependencies​

Focus on specific asset to understand dependencies:

msh ai context --asset revenue

Context Pack Structure​

Project Information​

  • Project ID, name, warehouse, default schema

Assets​

  • Asset metadata (ID, path, blocks)
  • Schema information
  • Transform SQL

Lineage​

  • Upstream dependencies
  • Downstream dependencies
  • Dependency types

Glossary Terms​

  • Business terms
  • Metrics
  • Dimensions
  • Policies

Schemas​

  • Flattened view of schemas per asset
  • Column types and constraints

Tests (Optional)​

  • Test definitions
  • Latest test statuses

History (Optional)​

  • Recent run history
  • Deployment history

Prerequisites​

  1. Manifest: Generate manifest before using context command

    msh manifest

  2. Glossary (optional): Create glossary for better context

    msh glossary add-term "Customer" --description "A customer entity"

Optimization​

Context packs are optimized for AI consumption:

  • Token limits respected
  • PII columns masked (based on policies)
  • Relevant information prioritized