AxiomCode Inc.
AxiomCode builds verifiable intelligence for AI software engineering. Its stated direction is to give agents and engineers formally grounded code context and deeper formal verification so they can understand what changes affect, execute tasks more completely, and ship safer software.
At a Glance
- Software engineers and engineering teams
- Teams using AI coding agents
- Enterprise and open-source software maintainers
- Security and code-verification teams
- +1 more
AI Tools by AxiomCode Inc.
(1)
AxiomCode Graph
Code Graph Engine for AI Agents
Discussions
No discussions yet
Be the first to start a discussion about AxiomCode Inc.
Latest News
AxiomCode Graph v0.2.3 released on GitHub
Published proof-of-technology case study on formal verification findings in Spring Cloud and Keycloak
Introduced AxiomCode Graph v0.1.0 with formal-methods code-graph and Defects4J benchmark results
Made AxiomCode Graph source available on GitHub under the FSL-1.1-Apache-2.0 license
Products & Services
An open-source code knowledge-graph builder that uses formal methods and language-aware rules to resolve calls, types, value flow, dependencies, configuration relationships, change impact, code paths, context, and affected tests. It is delivered as the @axiomcode/code-graph npm engine plus an agent-facing plugin/skill and MCP server.
AxiomCode's early-access verification engine. It consumes structured code context from AxiomCode Graph and applies connected analysis layers for deeper, AI-native formal verification of code behavior and intended constraints.
Market Position
AxiomCode positions its typed, formally grounded graph against syntax/tree-sitter-based code graph tools and grep-based search. In its published comparison it names GitNexus, CodeGraph, Graphify and Code-Review-Graph; it claims higher call-link coverage and 94.9% bug-level triggering-test coverage on its Defects4J evaluation, while emphasizing inspectable, compiler-style type resolution, value flow and explicit unresolved relationships.
Founding Story
AxiomCode says coding agents increasingly handle features, migrations, and fixes spanning whole codebases, but typically begin by grepping and repeatedly searching and rereading files. The company was started to provide a better foundation: a formally grounded, queryable graph that follows the program's actual structure, types, values, callbacks, and configuration rather than guessing from names, with AxiomEngine adding deeper verification on top.
Business Model
Revenue Model
AxiomCode Graph is publicly distributed as an open-source package under the FSL-1.1-Apache-2.0 license, while AxiomEngine is presented as an early-access product available through a demo/waitlist process. No public subscription or usage pricing was found.
Target Markets
- Software engineers and engineering teams
- Teams using AI coding agents
- Enterprise and open-source software maintainers
- Security and code-verification teams
- Developers working in Java, TypeScript, Python, JavaScript and C# codebases
- Helping AI coding agents understand unfamiliar repositories
- Change-impact analysis and safer implementation planning
- Selecting tests affected by a code change
- Explaining how code works and tracing call flows
- Finding bugs and analyzing behavior across connected code paths
- Reviewing diffs and checking whether changes are complete
- Spring Cloud / Spring Boot ecosystem
- Keycloak
- Kubernetes and OpenJDK are shown by AxiomCode as verification-in-practice targets