Gable

Overview

Gable sits in this category because of where it started, but it has repositioned. Its headline today is knowing where sensitive data flows before auditors ask, and the site's own title is code-level data flow lineage.

The mechanism is the interesting part and has not changed: static analysis of your source code, tracing data flows from consumer applications through backend systems, with no runtime agents, no sampling and no production overhead.

Data contracts survive as a feature rather than the headline. Gable drafts contracts from your existing data code and detects when a code change will violate one, and the API surface is still heavily contract-shaped.

The shift-left premise is that contracts belong with the engineers writing the data-producing code.

Its distinguishing claim against the rest of this category is where enforcement happens. Gable checks code, not landed data, so a violation surfaces in a pull request rather than in a warehouse the next morning.

Be clear about portability, because it is the material trade-off. Contracts are Gable-managed objects created through its own API, CLI or interface, not portable files you own, and the Open Data Contract Standard is not mentioned anywhere in its documentation.

Versioning exists at component and lineage level, with diffs between versions and lineage approvals that set a trusted baseline and track what has changed since.

Enforcement is a policy dial rather than a compatibility rule: contracts carry enforcement levels and a violation status, and an actions API drives responses.

Compliance is now an explicit vertical, with audit response kits, compliance exports and a dedicated page for BCBS 239, alongside a 2026 content programme dominated by AI governance.

It is trading and shipping, with a live careers page and SOC 2 Type 2 audited in June 2025, but it publishes almost nothing about itself: no founding year, headquarters, funding amounts, customer count or pricing.

Key features and capabilities

The same headings are used for every data contracts entry, so two tools can be read side by side.

What a contract covers

• Field-level data flow and lineage from application code through backend systems to data stores
• Data contracts over code-produced assets, with constraints and enforcement levels as API objects
• Sensitive and regulated data classification, with data tagging and compliance evidence
• Change impact and approvals, covering version changes, merge impact reports and baselines
• Quality expectations, service levels, pricing and terms are not published as contract sections

Specification and format

• The contract format is not published as a specification; there is no schema document
• Contracts are created and validated through the API and CLI rather than authored as a portable file
• The Open Data Contract Standard is not mentioned anywhere in the documentation
• Versioning exists at component and lineage level, with diffs between two versions
• No formal definition of a breaking change is published; enforcement levels are the nearest concept

How contracts are enforced

• Checking runs against source code through static analysis rather than against landed data
• Violations are detected during development, before deployment, which is the product's core claim
• Violation checks are published API operations, including checking lists of assets for compliance
• Consequences are configurable per contract through enforcement levels, with an actions API for responses
• The exact pull-request blocking mechanic and supported Git providers are not published

Authoring and review workflow

• Contracts are auto-drafted from existing data code rather than written from scratch
• Ownership is intended to sit with the software engineers writing the data-producing code
• Review happens through merge impact reports and lineage approvals against a trusted baseline
• Consumers register interest through contract subscriptions
• Onboarding is a documented project covering services, dependencies, field mappings and coverage

Tooling and validation

• A Python command-line tool installed with pip, authenticating by API key
• Contract inference from a data asset or a schema
• Lineage scanning, upload, export and generation for components, datastores and edges
• Compliance evidence exports for audits and examinations
• Linting a contract file, code generation and export to other formats are not published

Integrations

• S3 is the only data store named explicitly in the documentation
• The supported warehouse list is not published, though generic datastore objects exist
• An extensive published REST API covering contracts, components, lineage, actions and access control
• Package registry credentials imply pip and npm ecosystem coverage
• dbt, data catalogues and named continuous integration systems are not published

Where it runs and what it costs

• Managed software as a service only, running on Amazon Web Services
• No self-hosted option is published, and regions and data residency are not published
• A customer-side command-line tool runs in your environment and uploads results
• Gable states it collects only metadata, logs and metrics, not shared with external parties
• SOC 2 Type 2 audited in June 2025, with documented high availability and disaster recovery

Pricing

Price on requestQuote only; nothing published

No pricing is published at all. There is no pricing page, the URL returns an error and it does not appear in the site map. The published call to action is to request a demo, so it is entirely contact-sales. No free tier is published, no trial and no trial length, and the unit of pricing, whether per seat, per repository, per service or per data asset, is not published either. The sign-up page is not self-serve provisioning but a short qualification survey, so evaluation starts with a sales conversation.

Vendor pricing page →

Demos and videos

• Shifting left with data DevOps Webinar or talk · YouTube
• Gable documentation Tutorial · gable.ai

About Gable

Gable publishes remarkably little about itself. Its founders are named by first name and role only, Chad as chief executive, Adrian as chief technology officer, Daniel as founding engineer and James as chief product officer, with Chad Sanderson identified as chief executive through his bylines. No founding year, headquarters, headcount, customer count or funding amount is published; investors are named as Zetta, Crane, Databricks, B Capital, Capital One Ventures and Essence, and a careers page reference to Series A equity is the only stage signal. It is trading and hiring, with SOC 2 Type 2 audited in June 2025, and no acquisition or shutdown is published.

gable.ai