phodal/auto-dev
Auto Dev
auto-dev ist ein KI-natives Software-Engineering-Tool und eine Multi-Agenten-Entwicklungsplattform, die darauf ausgelegt ist, den gesamten Softwareentwicklungslebenszyklus zu automatisieren. Es fungiert als autonomer Orchestrator, der KI-gesteuertes Coding, Testen und Infrastrukturkonfiguration durch deklarative Agentenketten verwaltet. Das Projekt basiert auf einem Kotlin-Multiplatform-KI-Framework, wodurch Agentenlogik in verschiedenen Umgebungen und auf unterschiedlichen Geräteschnittstellen ausgeführt werden kann.
Die Plattform implementiert das Model Context Protocol, um Tools und Projektinformationen mit externen KI-Diensten auszutauschen. Sie zeichnet sich durch die Verwendung einer Retrieval-Augmented-Generation-Pipeline und baumbasiertem Code-Graphing aus, die abstrakte Syntaxbäume und Aufrufketten analysieren, um den Projektkontext zu komprimieren und Halluzinationen zu reduzieren. Eine interaktive Entwicklungsumgebung bietet Echtzeitsynchronisation von UML-Diagrammen, OpenAPI-Spezifikationen und Code-Diffs.
Die Funktionsbereiche decken autonome Softwareentwicklung ab, einschließlich dynamischer Aufgabenplanung, iterativer testgetriebener Reparatur und Migration von Legacy-Code. Das System handhabt zudem Infrastructure-as-Code-Automatisierung für Docker- und CI/CD-Konfigurationen, KI-gestützte Code-Reviews sowie die Koordination geteilter KI-Personas und Prompt-Spezifikationen über Teams hinweg.
Die Kernlogik ist in Kotlin Multiplatform implementiert, um eine konsistente plattformübergreifende Agentenbereitstellung sicherzustellen.
Features
- Autonomous Software Engineering - Functions as an autonomous agent system capable of generating source code, unit tests, and technical documentation.
- Multi-Agent Orchestrators - Orchestrates specialized AI agents and task chains to coordinate reasoning, planning, and code generation.
- Software Development Agents - Automates the complete software development lifecycle from planning and coding to testing and documentation.
- Agent Capability Extensions - Expands autonomous agent capabilities by integrating local or remote agents and defining specific intent actions.
- AI Agent Orchestrators - Coordinates specialized multi-agent workflows with custom prompts to complete complex development tasks.
- Multi-Agent Orchestration Platforms - Uses a multi-agent platform to automate the full development lifecycle from planning to deployment.
- AI Agent Integrations - Provides a dedicated execution language to connect and integrate external AI agents into the automation workflow.
- Model Context Protocol Integrations - Employs the Model Context Protocol as a standardized layer for exchanging tools and project information.
- AI Code Reviewers - Provides AI-driven analysis of code to identify bugs, suggest refactorings, and offer detailed review feedback.
- AI Software Engineering - Automates the complete software engineering lifecycle including code generation and legacy system refactoring.
- Autonomous Agent Orchestrators - Manages AI-driven coding, testing, and infrastructure via declarative agent chains and goal decomposition.
- Tree-Based Compression - Optimizes project context for complex refactoring tasks using a tree-based code graph engine.
- Retrieval-Augmented Generation - Uses retrieval-augmented generation to inject project-specific knowledge and historical context into AI interactions.
- Agent Behavioral Configuration - Tailors agent performance by configuring prompt templates and server settings via specialized instructions.
- Toolchain Context Integrations - Connects development toolchains and communication protocols to provide precise project information to AI models.
- LLM Provider Integrations - Integrates the development environment with various external LLM providers to power code generation and automation.
- Role-Based Model Assignment - Maps specific AI models to distinct functional roles such as reasoning, planning, or patch generation.
- Model Context Protocol Implementations - Implements the Model Context Protocol to standardize the exchange of tools and project information with AI services.
- Retrieval Augmented Generation Pipelines - Injects project-specific knowledge and historical insights into AI prompts via a retrieval-augmented generation pipeline.
- Project-Specific Overrides - Customizes code and commit message generation by replacing system-level prompts with project-specific files.
- Agent Configurations - Implements local YAML and prompt-based configuration files to define agent capabilities and task chains.
- AI Code Generation Workflows - Uses specialized AI agents and context-aware workflows to automate the creation of application code and queries.
- Feedback-Loop Pipelines - Implements a pipeline that uses build logs and test results to iteratively refine AI prompts and repair code.
- Test-Driven Automated Repairs - Generates unit tests and automatically repairs source code based on specific test failures.
- Development Workflow Automation - Orchestrates complex coding and building sequences by combining AI reasoning with manual user adjustments.
- Kotlin Multiplatform Development - Implements a cross-platform core that allows agent logic to run across diverse environments and interfaces.
- AI Prompt Configurations - Manages and overrides prompt configurations to distribute consistent coding patterns across a team.
- AI Agent Frameworks - Provides a cross-platform AI framework built on Kotlin Multiplatform for consistent agent deployment.
- Multiplatform Agent Runtimes - Uses a Kotlin Multiplatform core to ensure AI agent logic runs consistently across diverse environments.
- Automated Software Testing - Automatically creates and executes unit tests while iteratively fixing failing code to ensure system stability.
- Test-Driven Repair Loops - Generates unit tests and iteratively repairs source code based on test failures to ensure stability.
- Scenario-Based Model Selection - Optimizes output by setting specific model types for different development scenarios and tasks.
- Enterprise Agent Tailoring - Customizes AI agents and private models to align with individual, team, or enterprise workflow requirements.
- Code Explanation - Troubleshoots errors and provides natural language explanations of source code logic and smells.
- Code Validation Pipelines - Provides an automated pipeline combining testing and static analysis to validate AI-generated code reliability.
- Context-Aware Code Generators - Analyzes the project state to synthesize context-aware SQL queries and web pages.
- Model Context Protocol Clients - Implements the Model Context Protocol to exchange tools and project information with external AI services.
- Prompt Optimization - Uses verification results to automatically refine AI prompts and update project knowledge bases.
- AST-Based Code Graphs - Analyzes AST and call chains to compress project context for efficient large-scale refactoring.
- Code Refactoring Tools - Identifies code smells and performs intelligent structural improvements to modernize existing source files.
- Environment-Aware Configurations - Creates Dockerfiles and CI/CD files tailored to specific target environments and build tools.
- Interactive Code Artifact Previewers - Provides a rendering system for real-time synchronized previews of UML diagrams, diffs, and OpenAPI specifications.
- Secure Command Wrapping - Executes AI-generated shell and SQL commands within a secure, wrapped environment to prevent system risks.
- Cross-Platform Deployments - Runs AI agent logic across diverse operating systems using a unified cross-platform core.
- Infrastructure as Code Tools - Produces Dockerfiles and CI/CD pipeline configurations using a machine-readable, declarative approach.
- Infrastructure Generators - Provides automated generation of Dockerfiles and CI/CD configurations based on project characteristics.
- Codebase Blueprinting - Analyzes codebases using AST and call-chain tracing to generate business logic maps and architectural blueprints.
- Enterprise Agent Orchestration Platforms - Creates specialized agents using private models tailored for enterprise-scale workflow management.
- Intelligent Task Planning - Generates dynamic task lists that adjust based on project context to track and manage development progress.
- Legacy Software Modernization - Analyzes legacy codebases to automatically translate them into modern architectures with verification tests.
- AI-Driven Error Resolution - Captures test failures and build logs to automatically diagnose and resolve errors using AI agents.
- AI Test Intent Specifications - Defines high-level test intents and automatically runs fixes to maintain high code quality.
- Static Analysis - Integrates automated static analysis into the code generation pipeline to identify bugs and quality issues.
- Artifact Synchronization Canvases - Provides an interactive canvas for reviewing patches, diffs, and verifying compilation in a web view.
- Interactive Development Canvases - Renders and synchronizes diagrams, OpenAPI specifications, and UI prototypes using an interactive canvas.
