act101

act101 is a revolutionary developer tool that fundamentally changes how AI coding agents interact with source code. It is the first dev tool that enables AI agents to perform language-aware code operations, including refactoring, porting, and analysis across an unprecedented 163 programming languages. Unlike traditional AI coding assistants that rely on whole-file rewrites, act101 exposes a comprehensive set of Abstract Syntax Tree (AST) operations to agents through the Model Context Protocol (MCP). This allows agents to execute precise, typed operations like extract-function, rename-symbol, or move-module with cross-file consistency and automatic checkpointing. The tool is built as a single native Rust binary with no plugin runtime, no package graph, and no supply chain attack surface. It operates entirely on the user's machine with no telemetry or data collection, ensuring complete code privacy. act101 is designed for professional developers using AI coding agents like Claude Code, Cursor, Codex, and OpenCode who need to perform sophisticated code transformations that go beyond simple text generation. By providing 183 AST refactor operations, 30 codebase analyzers, 15 query operations, 8 porting operations, and 10 pre-built agent skills, act101 empowers developers to automate complex engineering workflows with unprecedented precision and safety.

183 AST-Aware Refactor Operations

act101 provides AI agents with a comprehensive library of 183 typed, AST-aware refactor operations that go far beyond simple find-and-replace. These operations include extract-function, rename, move-symbol, inline, convert-to-dataclass, extract-trait, add-type-hints, generate-init, organize-imports, and 174 more. Each operation is executed at the syntax tree level, meaning the agent understands the actual structure of the code, not just its text. This enables cross-file consistency, automatic checkpoint creation before every operation, and instant undo if the result is unsatisfactory. The operations work across all 163 supported grammars, from Python and Rust to COBOL and Elixir.

8 Cross-Language Porting Operations

act101 introduces a sophisticated state-machine approach to cross-language code migration through 8 dedicated porting operations. The process begins with port_contract to anchor the source-to-target migration, followed by port_inventory to enumerate every symbol that must move, port_order to resolve dependency ordering, and a port_manifest state machine with init, add, update, remove, and note operations. This systematic approach enables agents to port code between any two of the 163 supported grammars, such as C to Rust, Ruby to Elixir, or COBOL to Java, with full awareness of language-specific idioms and patterns.

30 Codebase Analyzers

The tool includes 30 sophisticated codebase analyzers that provide AI agents with deep structural understanding of the code before any changes are made. These analyzers cover cohesion, coupling, cycles, chokepoints, hotspots, dead code, layers, seams, clusters, surface area, fan balance, migration readiness, type completeness, and more. Each analyzer returns structured data that agents can use to make informed decisions about refactoring priorities, identify architectural issues, and plan migrations. This structural mapping capability transforms how agents approach complex codebases, replacing guesswork with data-driven insights.

10 Pre-Built Agent Skills

act101 ships with 10 pre-built agent skills that compose individual operations into common engineering workflows. These skills include architecture-audit, code-review, refactoring, code-navigation, code-generation, migration-assessment, boundary-analysis, change-impact, health-check, and architectural-refactoring. Each skill can be invoked with a simple command like /skill-name in Claude Code, Cursor, or other MCP-compatible clients. These skills dramatically reduce the complexity of orchestrating multiple operations, allowing developers to accomplish sophisticated tasks with a single command while maintaining full control over the process.

Automated Large-Scale Code Refactoring

Development teams can use act101 to automate complex refactoring tasks across entire codebases with precision and safety. For example, a team maintaining a large Python monolith can instruct an AI agent to extract functions from overlong methods, rename symbols across multiple modules for consistency, convert classes to dataclasses, and reorganize imports, all with automatic checkpointing and instant undo. The agent uses the 183 AST-aware operations to make surgical changes that preserve comments, formatting, and cross-file dependencies, eliminating the risks associated with whole-file rewrites.

Cross-Language Code Migration

Organizations modernizing legacy systems can leverage act101 for systematic code porting between languages. A team migrating a C library to Rust can use the porting state machine to create a contract, inventory all symbols, resolve dependency ordering, and track progress through the manifest. The agent handles the nuanced translation of language-specific patterns, such as converting C pointers to Rust references or Ruby dynamic typing to Elixir pattern matching. This structured approach ensures completeness and correctness while dramatically reducing the manual effort required for language migration.

Architectural Analysis and Health Checks

Engineering leaders can use act101 to gain deep insights into their codebase architecture without manual analysis. By invoking the architecture-audit or health-check skills, they can automatically identify coupling issues, dependency cycles, dead code, hotspots, and boundary violations across any of the 163 supported languages. The 30 analyzers provide actionable data that teams can use to prioritize technical debt, plan refactoring sprints, and enforce architectural standards. This capability is particularly valuable for onboarding new team members or assessing code quality before major feature work.

AI-Assisted Code Review

Development teams can integrate act101 into their code review workflow to automate structural analysis of proposed changes. When reviewing a pull request, an AI agent can use the code-review skill to analyze the change impact, identify potential coupling issues, check for migration readiness, and verify architectural boundaries. The agent can provide detailed feedback on structural concerns that would be difficult or time-consuming for human reviewers to catch, such as unintended dependency cycles or violations of layering rules. This enhances the review process without replacing human judgment on design decisions.

How does act101 differ from traditional AI coding tools?

Traditional AI coding tools operate on text, treating code as a string to be generated or replaced. This leads to whole-file rewrites that often lose comments, break formatting, and provide no undo capability. act101 operates on the Abstract Syntax Tree (AST), giving AI agents typed, structural operations that understand the actual meaning of the code. This enables surgical changes with cross-file consistency, automatic checkpointing, and instant undo. Additionally, act101 runs entirely on the user's machine with no telemetry, ensuring code privacy, and supports 163 languages in a single native binary with no dependencies.

Is act101 safe to use with proprietary code?

Yes, act101 is designed with security and privacy as core principles. The tool runs as a single native Rust binary entirely on the user's machine. It performs no indexing, no caching, and no data collection. Your code never leaves your machine. The only external communication is for license verification, which may contact act101 servers. There is no telemetry, no plugin runtime, no package graph, and no supply chain attack surface. This makes act101 suitable for use with even the most sensitive proprietary codebases.

What AI coding agents are compatible with act101?

act101 is built on the Model Context Protocol (MCP), making it compatible with any MCP-aware client. This includes Claude Code, Cursor, Codex, OpenCode, and Windsurf. The tool exposes its operations through a built-in MCP server that agents can connect to directly. Developers using other AI coding tools can integrate act101 as long as those tools support the MCP standard. The 10 pre-built agent skills are designed to work seamlessly with these clients, allowing developers to invoke complex workflows with simple commands like /skill-name.

Do I need to install any additional runtimes or dependencies?

No, act101 is distributed as a single native Rust binary with no external dependencies. There is no need to install Python, Node.js, Java, or any other runtime. The binary contains all 163 grammars, all operations, and the MCP server built-in. This zero-dependency approach eliminates compatibility issues, reduces attack surface, and ensures consistent behavior across different operating systems. Simply download the binary, run it, and connect your AI agent to start using all features immediately.