Skip to content
Best Data Catalogs mark: an unfinished lineage pathBest Data Catalogs

Platform review

Semantic layer

AtScale review: certified metrics for AI agents over MCP

Short answer

AtScale is a universal semantic layer whose MCP server gives agents certified metrics by name, with per-agent OAuth identity and row- and column-level security. It scores 5.87 and ranks 6 of 9. It does not cover lineage across the stack, and its prices are not published.

By The Lineage Desk · Best Data Catalogs · Published

Desk research from public vendor pages. Editorial assessment.

5.87 / 10

Rank 6 of 9 on default weights. Change the weights

What does AtScale do for AI agents?

AtScale is a universal semantic layer for AI and analytics. Its MCP server lets agents ask for certified metrics by name, while queries compute inside your data platform.

Agent interface
AtScale MCP server (list_models, explore_columns, focus_columns, get_sml_skills, run_query, get_outbound_queries); AtScale Ava for chat users
Lineage
Not described (a semantic layer, not a catalog)
Deployment
Queries compute inside your data platform; hosting options are not stated on the pages we read.
Best for
Enterprises that need certified metrics served to many agent clients under per-agent security

Which hops from warehouse to agent does AtScale document?

AtScale context path: documents 4 of 5 stages.1Warehouse and lakehouse2Transformation (dbt, ETL)3BI and semantic logic4Governance and access rules5Delivery to agents (MCP or API)
  • Solid dotDocumented on the vendor's pages
  • Dashed hollow dotNot documented on the pages we read

  1. 1Warehouse and lakehouse: Documented

    Queries compute in your data platform.

    Source: AtScale: semantic context for AI via MCP

  2. 2Transformation (dbt, ETL): Not documented

    Transformation tools are not described on the pages we read.

  3. 3BI and semantic logic: Documented

    Semantic layer for analytics and AtScale Ava.

    Source: AtScale homepage

  4. 4Governance and access rules: Documented

    Per-agent OAuth identity, row- and column-level security.

    Source: AtScale: semantic context for AI via MCP

  5. 5Delivery to agents (MCP or API): Documented

    AtScale MCP server.

    Source: AtScale: semantic context for AI via MCP

A filled node means the vendor's own pages describe that hop. It does not measure how well it works.

What are AtScale's key features?

What does AtScale claim?

What does AtScale cost?

Not published Consumption pricing by deployed semantic objects, in Standard and Enterprise editions, with no per-seat or per-query fees. No prices are published.

Entry price: Not published

Source: AtScale pricing

How does AtScale score on each criterion?

CriterionWeightScoreWhy
Agent access to context15%9 / 10An MCP server with named tools gives agents certified metrics by name; seven agent clients are named.

Source: AtScale: semantic context for AI via MCP

Cross-stack lineageTied lowest13%2 / 10A semantic layer, not a catalog; cross-stack lineage is not described.

Source: AtScale homepage

Automated context buildTied lowest13%4 / 10Describes a 30-day workshop from kickoff to results; automated discovery is not described.

Source: AtScale: semantic context for AI via MCP

Governance agents can use13%8 / 10Per-agent OAuth identity, row- and column-level security and a log of every call.

Source: AtScale: semantic context for AI via MCP

Business definitions and metricsTied highest12%9 / 10A universal semantic layer serving certified metrics to BI and agents.

Source: AtScale homepage

Answer quality checks8%5 / 10Every call is logged, and AtScale cites a preliminary result of twice the accuracy on BIRD-Interact.

Source: AtScale: semantic context for AI via MCP

Integration breadthTied lowest10%5 / 10Agent clients are listed, but no data source count was read.

Source: AtScale: semantic context for AI via MCP

Pricing transparency8%4 / 10The pricing model is published (deployed semantic objects, no per-seat or per-query fees) but no prices.

Source: AtScale pricing

Deployment options8%5 / 10Queries compute inside your data platform; hosting options are not stated.

Source: AtScale: semantic context for AI via MCP

Where does it score well, and where does it fall short?

Scores well

Highest in set on: Semantics.

  • Per-agent identity with row- and column-level security for agent queries.
  • Many named agent clients.
  • A pricing model without per-seat or per-query fees.

Falls short

Lowest in set on: Lineage, Automation and Integrations.

  • No catalog and no lineage across the stack.
  • No prices are published.
  • The stated path to results is a 30-day workshop, slower than the day-one or same-day starts the context platforms describe.

Key facts

Company
Not covered in our sources.
Customers named
Named on AtScale's pages: Vodafone Portugal (with Celfocus) and Blue Yonder.

Source: AtScale: semantic context for AI via MCP

How does AtScale compare head to head?

AtScale alternativesBuild a shortlist with AtScale

Frequently asked questions

What does AtScale cost?

Not published. Consumption pricing by deployed semantic objects, in Standard and Enterprise editions, with no per-seat or per-query fees. No prices are published.

Does AtScale have an MCP server?

Yes. The AtScale MCP server exposes tools such as list_models and run_query.

Can AtScale be self-hosted?

Queries compute inside your data platform; hosting options are not stated on the pages we read.