Open-source alternatives to Typeshed

Pyrefly is a static type checker for Python that operates as a language server, delivering real-time diagnostics, completions, and navigation in any editor supporting the Language Server Protocol. It also performs static tensor shape analysis, using symbolic dimension variables and arithmetic to verify shape consistency in deep learning models without runtime execution. Beyond core type checking, Pyrefly supports gradual adoption workflows: it can generate a baseline of known errors so only new issues are reported, migrate configuration from other type checkers, and automatically suppress exi

This project is a high-performance static type checker and comprehensive development toolkit for Python. It functions as a core analysis engine that identifies type inconsistencies and enforces code correctness, while simultaneously providing a language server implementation to deliver real-time diagnostics and intelligence directly within development environments. The tool distinguishes itself through a parallelized execution engine that maximizes performance across large-scale codebases and monorepo structures. It supports gradual type adoption, allowing developers to integrate type checkin

MonkeyType is a runtime type inferencer and analysis tool for Python. It collects actual argument and return types from function calls during program execution to automate the production of static type annotations and compatible stub files. The system converts captured execution traces into static type hints, which can be inserted directly into source code function definitions or used to generate external type stubs. This process allows for the addition of type safety to existing codebases by analyzing real-world data types rather than through manual auditing. The tool's capabilities cover r

This project is a comprehensive guide and educational resource for the TypeScript language. It covers the fundamental principles of the language, including its structural type system, static type analysis, and the process of transpiling typed source files into JavaScript. The material details how to model complex data and reusable type logic using generics, conditional types, and mapped types. It also explains the use of declaration files to provide type safety for external JavaScript libraries and the integration of type checking into existing JavaScript projects via JSDoc annotations. The

lua-language-server is a static analysis tool and type checker for the Lua language that implements the Language Server Protocol. It provides a system for detecting type mismatches, unused code, and logic errors in source files. The project features an inference-based type system that deduces variable types and supports optional annotations and meta files to enforce type safety. It allows for the definition of custom types and function signatures, including support for overloads and member visibility controls. The server provides a broad set of IDE capabilities, including real-time code auto

This project is a JavaScript style guide and a set of coding standards designed to ensure codebase consistency. It provides a collection of best practices for writing maintainable and idiomatic code, focusing on variable naming, type checking, and execution context. The project serves as a static analysis guide, offering frameworks and guidelines for using linters and automated tools to identify bugs and verify code correctness before deployment. It establishes standards for formatting, syntax, and variable declarations to standardize code style across a project. The scope of the project cov

Pyre is a high-performance static type checker and analysis tool for Python. It identifies type errors and ensures type safety without executing the program, utilizing a static type inference engine to maintain consistency across functions. The project is distinguished by an incremental type analysis engine that operates as a background daemon. This system monitors filesystem changes to re-validate only modified parts of a project, reducing the time required for repeated analysis. It also includes a static analysis security tool that uses taint analysis to track untrusted data flows and ident

Monty is a sandboxed execution environment designed primarily for running Python code generated by AI models. It provides a secure, isolated runtime that blocks host access, enforces resource limits, and supports pre-execution type checking against built-in type hints to catch signature mismatches before code runs. The sandbox can persist its interpreter state at external function calls, allowing execution sessions to be serialized, stored, and later resumed from a file or database. What distinguishes Monty is its combination of stateful, resumable execution with multi-language native embeddi

PMD is a multi-language static code analyzer used to identify programming flaws, unused variables, and dead code without executing the program. It functions as a code smell detector and coding standard enforcer, ensuring source code adheres to specific naming conventions, structural requirements, and project style guides. The project features an XPath-based rule engine that allows users to define custom analysis patterns using queries against an abstract syntax tree. It also includes a copy-paste detector to identify duplicated code blocks across multiple files and a visual rule designer for

Checkstyle is a Java static analysis tool and linter designed to identify and enforce coding standards and best practices. It functions as a code quality auditor and Javadoc validation tool, checking source code against configurable rulesets to ensure structural and stylistic consistency. The project allows for the creation of custom linting rules by extending a core API to inspect the abstract syntax tree. It further enables specialized validation through the use of XPath expressions to query the syntax tree for specific code patterns and violations. Capability areas include the enforcement

Infer is a static analysis toolset for Java, C, C++, and Objective-C designed to detect memory leaks, null dereferences, and resource bugs. It functions as a multi-language bug finder that identifies race conditions, deadlocks, and memory safety issues by translating source code into a common intermediate representation for analysis. The project distinguishes itself through an inter-procedural data flow analyzer that tracks movement between sources and sinks to detect tainted flows and generate data flow graphs. It also includes a framework for verifying temporal properties and reachability u

p3c is a Java static analysis tool and code quality linter designed to enforce professional coding guidelines and quality standards. It utilizes a set of custom rules based on the PMD engine to scan source code for style violations, performance bottlenecks, and potential bugs. The project is distributed as an IDE linting plugin that provides real-time feedback and warnings during development. It also includes functionality for pre-commit code quality gates, allowing modified files to be scanned and blocked if they violate defined rules before being committed to version control. The analysis

Error Prone is a static code analyzer and Java compiler extension that identifies common programming mistakes during the build process. It functions as a compiler wrapper that flags potential errors as compile-time failures to prevent bugs from reaching execution. The tool integrates directly into the Java compilation workflow to provide compile-time validation. It allows for the definition of custom linting rules and analysis checks to enforce specific coding standards and detect prohibited API usage. The system utilizes abstract syntax tree analysis and type-aware pattern matching to inspe

This project is a comprehensive set of architectural and coding standards for organizing and maintaining high-quality JavaScript applications. It provides a framework for JavaScript project best practices across the full software development lifecycle, establishing unified guidelines for project organization and development. The guidelines cover specific standards for REST API design, utilizing resource-oriented interfaces and standardized HTTP methods. It also includes a web accessibility standard focused on semantic HTML and automated audits, alongside a defined Git workflow standard for br

Pylint is a static code analyzer for Python that scans source code for errors, coding standard violations, code smells, and type-related issues without executing the program. It functions as a plugin-based linter framework, allowing users to extend its analysis capabilities with custom or third-party checks for project-specific rules and framework support. The tool also includes a duplicate code detector that identifies identical or near-identical code blocks across a project to help reduce redundancy. Beyond its core linting functionality, Pylint can generate UML class and package diagrams f

Flake8 is a Python style checker and static analysis tool that identifies programmatic errors and violations of coding standards in source files. It serves as a linting wrapper that aggregates multiple checkers into a single interface, functioning as a pluggable framework for Python code analysis. The tool incorporates a cyclomatic complexity analyzer to measure the logical complexity of functions and identify code that requires refactoring. It is designed for extensibility, allowing the integration of custom plugins and rules through defined entry points. The system provides capabilities fo

Credo is a static analysis tool and linter for Elixir. It functions as a code quality analyzer that scans source code to identify stylistic inconsistencies, common mistakes, and potential security vulnerabilities. The tool provides a customizable framework for defining and testing specialized rules to enforce project-specific coding standards. It identifies complex code fragments and duplication to highlight opportunities for refactoring and simplification. Its capabilities cover automated code reviews, the enforcement of Elixir coding standards, and real-time developer feedback through edit

NullAway is a Java static analysis tool and build-time detector designed to identify null pointer risks. It functions as a nullability checker that uses annotations to verify that references are not null before they are dereferenced, preventing runtime crashes. The analyzer implements the JSpecify standard to ensure consistent nullability annotations across different Java libraries. It distinguishes itself through the use of a service provider interface to model the nullability behavior of third-party libraries that lack source annotations and by providing specialized support for Lombok-gener

code2flow is a static program flow mapper and source code call graph generator. It analyzes source code to produce visual flow diagrams that map function call relationships and execution paths. The project includes an asynchronous call trace visualizer that follows execution paths through async and await calls to map the logic of asynchronous programs. It also provides a programmable code analysis interface, allowing the call graph generation logic to be integrated into other software applications. The system handles static code analysis by converting source code into abstract syntax trees t

This project is a static analysis linter, code quality tool, and language auditor for Rust. It functions as an automated refactoring system designed to identify common mistakes and suggest idiomatic improvements for Rust source code. The tool identifies non-idiomatic patterns, performance bottlenecks, and code smells to improve the overall correctness and quality of the code. It specifically audits memory safety by flagging suspicious use of unsafe blocks and pointer manipulations and detects inefficient operations to optimize execution speed. The analysis surface covers coding style enforce

DefinitelyTyped is a community-maintained type store and centralized JavaScript type registry. It serves as a repository of static TypeScript type declarations for third-party JavaScript libraries, providing the necessary metadata to enable compile-time safety and editor intelligence for external modules not originally written in TypeScript. The project operates as a collaborative ecosystem where contributors define, validate, and maintain type declarations through a structured review process. This involves mapping type definition versions to specific library and compiler releases to ensure s

Reason is a strongly typed functional language and multi-target compiler designed to produce either JavaScript for web environments or native machine code for operating systems. It serves as an OCaml-compatible language that shares an abstract syntax tree with OCaml, providing a practical syntax for the OCaml ecosystem. The project functions as a type-safe frontend language, incorporating JSX support to build user interfaces through structural elements that translate into type-safe function calls. It enables deep integration with the JavaScript ecosystem, allowing the use of existing librarie

Pylint is a static code analysis tool for Python that checks source code for errors, coding standard violations, and code smells without executing the program. It parses code into an abstract syntax tree and walks the tree to detect issues, enforces configurable style rules and naming conventions, and identifies duplicate code blocks by comparing tokenised source sequences. The tool also includes an inference engine that deduces variable types by simulating code paths, enabling deeper analysis even in untyped code. What distinguishes Pylint is its plugin-based checker architecture, which allo

PHP-Parser is a tool that converts PHP source code into an abstract syntax tree for static analysis and programmatic manipulation. It functions as a parser, a code generator, and a static analysis framework. The project enables the programmatic construction of abstract syntax tree nodes through a fluent interface and provides the ability to transform these trees back into formatted source code. It includes a serializer that exports abstract syntax trees to JSON format and reconstructs them from strings. The toolset covers several capability areas, including namespace resolution, constant exp

Mago is a PHP static analysis tool, linter, and code formatter. It detects type errors, dead code, and logic bugs through advanced type inference and flow analysis, while providing a deterministic engine to rewrite source files according to style guides. The project distinguishes itself with a WebAssembly runtime that enables PHP analysis to execute entirely within a web browser, allowing results to be shared via URL. It also provides a minimal Dockerized toolchain distributed as a scratch image for consistent execution across different processor architectures. The toolset covers architectur

Checkov is a static analysis tool and security scanner designed to identify misconfigurations in infrastructure as code, container images, and Kubernetes configurations. It functions as a cloud security posture tool, an SCA vulnerability scanner, and a secret scanning utility to prevent security breaches and version control leaks. The project distinguishes itself through deep graph analysis and variable resolution, allowing it to map relationships between interconnected resources and evaluate the final state of infrastructure attributes. It provides extensibility for defining custom security

ApplicationInspector is a multi-language static analysis tool designed to detect specific features and characteristics within source code. It utilizes a declarative JSON rules engine to identify patterns and structural tags across project directories without requiring the analyzer to be recompiled. The system distinguishes itself through a code version differ that compares two different source paths to report changes in detected features. It also provides utilities for creating and validating custom JSON-based rules, including a validation pipeline to verify syntax and identifier uniqueness.