Reasonix is a local AI coding agent engine written in Go, providing a terminal UI, desktop application (Electron + React), and HTTP server for interacting with large language models (OpenAI, Anthropic, DeepSeek, etc.) to automate software engineering tasks. The kernel manages sessions, context windows, planning, sub-agent delegation, tool execution (file operations, shell, browser, MCP servers), and a sophisticated safety sandbox. It persists conversation history with checkpoints for undo/rewind, tracks token usage, and maintains an evidence ledger for auditing tool effects. The architecture is a monolithic Go binary with multiple frontends: a CLI/TUI, a desktop host that communicates via SSE/JSON, and a REST API server. A plugin system using the Model Context Protocol (MCP) allows extension. Separate Cloudflare Workers handle account management, forum, and crash reporting. The codebase is well-structured with a clean layered architecture (base utilities, domain contracts, state persistence, runtime agent, assembly wiring, frontends), extensive test coverage, and internal linters enforcing quality. It is actively maintained with frequent commits and professional CI/CD pipelines.
This repository is a comprehensive, multi-package AI agent framework developed by LangChain, known as 'Deep Agents'. It provides a production-ready SDK ('deepagents') built on LangGraph for creating stateful, middleware-driven agents with capabilities such as filesystem manipulation, shell execution, memory, sub-agents, skills, human-in-the-loop, and summarization. The primary client is a high-quality CLI/TUI application ('deepagents-code') that offers an interactive coding assistant with a rich Textual terminal interface, plugin system, MCP OAuth integration, and sandboxed execution via multiple providers (Daytona, Modal, Runloop, Vercel). A deployment CLI ('deepagents-cli') manages agent lifecycle on LangSmith, and an experimental persistent runtime ('talon') supports cron jobs, voice, and channels like Telegram/WhatsApp. A comprehensive evaluation suite ('evals') includes benchmarks (FRAMES, Nexus, BFCL, tau2, MemoryAgentBench, Context-Bench, DRBench) with LLM-as-judge, trajectory scoring, and LangSmith reporting. The architecture emphasizes modularity: middleware stack, hooks system, backend protocols, and profile system. While primarily an application, the evals and dataset curation give it a hybrid character.
planning-with-files is a persistent, file-based task-planning system for AI coding agents. It maintains task_plan.md, findings.md, and progress.md on disk and wires into agent lifecycle hooks (SessionStart, UserPromptSubmit, PreToolUse, PostToolUse, PreCompact, Stop) across Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, Pi, Hermes, and a dozen other assistants, re-injecting plan context into the model's context window on every turn. Plans survive context loss, /clear, compaction, and crashes; a session-catchup mechanism recovers prior transcripts. A v3 gated mode can block the agent from stopping until plan phases complete, backed by an append-only JSONL run ledger, stall detection, and SHA-256 plan attestation that makes hooks refuse to inject tampered content. Architecturally it is a distribution hub rather than a single application: a canonical skill under skills/planning-with-files is mirrored byte-identical into 15+ IDE-specific directories, with native TypeScript plugins (OpenCode, Pi, DSH), a Python plugin (Hermes), and shell/PowerShell hook dispatchers. Plan resolution is a fail-closed chain (PLAN_ID env → .active_plan pointer → newest plan dir → legacy root) guarded by slug validation, containment checks, symlink/junction rejection, and strict PWF_PLAN_ROOT pinning. The engineering is unusually rigorous for an agent-skill repo: 400+ pytest tests, byte-parity tests between shell/PowerShell/Python/TypeScript twins, a PATH-stub BSD userland simulation to catch GNU-only flags that silently fail on macOS, and heavy Windows PowerShell 5.1 / Git Bash portability hardening. Non-obvious patterns include KV-cache-stable injection output (no timestamps or free text in summaries, wall-clock normalization to T00:00:00Z) and an mtime-keyed attestation cache for performance. The repository is a developer tool, not a data/research project; although it ships benchmark and eval documentation (assertion pass rates, recovery-turn comparisons), those are verification evidence for the skill, not analysis artifacts. It is actively maintained at v3.24.0 with recent commits, 4 open issues, and 100+ releases; technical drift is low, with the main maintainability risk being deliberate duplication across 20+ canonical mirrors, mitigated by sync-ide-folders.py and parity tests but still a possible source of silent drift.
Qwen Code (formerly Gemini CLI) is an open-source, multi-platform AI coding assistant designed to operate in the terminal, web shell, desktop (Tauri), VS Code, JetBrains, and chat channels (DingTalk, WeChat, Slack, etc.). The architecture is a monorepo with ~30+ packages centered around a core library (`packages/core`) that provides agent runtime, tool orchestration, memory, MCP support, and telemetry. The primary entry points are the CLI (`packages/cli` with an Ink-based TUI), the web shell (`packages/web-shell`), and a daemon server (`packages/cli/src/serve/`) that exposes REST/SSE endpoints for session management, file editing, and sandboxed shell execution. The system supports multiple LLM providers (OpenAI, Anthropic, Gemini, DeepSeek, etc.) and features sub-agents, goal-oriented workflows, approval modes, browser automation (CUA driver in Rust), and a Java managed-agent server for enterprise deployments. Data flows are user-to-agent: user input enters via a client (CLI, web, IDE), is processed by the core agent loop, which invokes tools (shell, file system, LSP, MCP servers) and streams responses back. External dependencies include Node.js v22+, pnpm, and platform-specific native binaries (Rust CUA driver, Electron desktop). The codebase is actively maintained with daily commits, extensive integration tests, and CI/CD pipelines. While it is a software application, it includes telemetry modules and a few Python scripts for benchmark analysis, but no research-grade data pipelines or ML training code.
Codewhale is an open-source, terminal-native AI coding agent that operates as a TUI (ratatui), CLI, headless runtime, MCP server, and a web frontend. It manages full development workflows: reading repositories, editing files, running commands, reviewing changes, and orchestrating multi-agent fleets. The codebase is a Rust monorepo with ~52 crates, a Next.js website with 18+ locales, a VS Code extension, packaging for npm/Homebrew/Arch/Windows, and companion pet apps. The architecture is event-driven with a pluggable tool system, model-agnostic provider routing, session-based journaling, and a contract-based command framework. It supports dozens of LLM providers (DeepSeek, OpenAI, Anthropic, Ollama, etc.) and features sandboxed execution, telemetry with a strict no-IP privacy contract, and a plugin marketplace. The project is heavily tested with unit, integration, PTY/Cucumber, and hermetic bash tests, and enforces budgets for dead code and blocking calls.
PiDeck is an open-source Electron desktop application that serves as a comprehensive workbench for AI coding agents, primarily wrapping the `pi` coding agent and the DeepSeek Harness (DSH) backend. It provides a rich session-centric chat workspace with features including multi-project management, Git integration, a built-in terminal, voice transcription, image generation, scheduled automation tasks, a desktop pet companion, and extensive configuration options. The application supports importing sessions from other tools like Claude, Codex, Cursor, and OpenCode, and offers a web service variant for remote access. Architecturally, PiDeck follows the standard Electron pattern with a main process (Node.js services, IPC handlers, filesystem operations), a preload script for secure IPC bridging, and a renderer process built with React 19, TypeScript, Jotai for state management, and Tailwind CSS v4 with shadcn/ui-style components. The codebase is organized into `src/main/` for backend services, `src/renderer/` for the UI, and `src/shared/` for cross-process types and utilities. Key entry points include the main process index, renderer bootstrap, and numerous IPC modules that validate inputs and delegate to domain services. Data flows center on session files stored in JSONL format (for pi) or compressed JSONL (for DSH), with model catalogs, usage statistics, and configuration persisted in user data directories. The application manages multiple backends through a unified session catalog, with runtime state synchronized via IPC events and Jotai atoms. External dependencies include the pi-coding-agent CLI, DSH runtime packages, node-pty for terminal emulation, whisper.cpp for local speech-to-text, and electron-updater for cross-platform updates. The codebase demonstrates strong engineering practices with extensive unit tests (node:test), Playwright E2E tests, pure-function modules extracted for testability, and consistent IPC boundary validation patterns.
OpenEvolve is an open-source evolutionary coding agent that uses large language models (LLMs) to iteratively improve programs. It implements a MAP-Elites quality-diversity algorithm with island-based evolution, supporting multiple LLM backends (OpenAI, Claude Code, etc.) and a cascade evaluation pattern. The system accepts a user-provided initial program and an evaluator script, then repeatedly mutates the program via LLM-generated SEARCH/REPLACE diffs or full rewrites, evaluates the fitness, and maintains a diverse population across feature dimensions. The core architecture is a library with a CLI entry point (`openevolve-run`), a high-level API (`run_evolution`, `evolve_function`, `evolve_code`), and a controller that orchestrates the evolution loop, prompt sampling, LLM inference, parallel evaluation, and database storage. The repository includes 20+ diverse examples ranging from algorithm optimization (AlgoTune, TSP, circle packing) to mathematical discovery (Heilbronn, kissing number) and hardware-specific kernel tuning (MLX Metal, MLIR). The codebase is well-structured with extensive unit and integration tests, typed configuration, and a modular design that separates concerns (evaluator, LLM, database, prompt templates). It is actively maintained with frequent commits and responsive issue triage.
A production-grade AI coding agent (the Codex CLI) that operates as a terminal-based assistant. It provides a multi-modal, extensible platform for thread-based conversations with language models (OpenAI, Bedrock, Ollama), including sandboxed code execution, MCP tool integration, plugin/skills/hooks extension system, and state persistence via SQLite. The architecture is a Rust monorepo with a client-server split: a high-level TUI or CLI sending JSON-RPC messages to an app-server that manages session state, model routing, and execution policies. The system supports cross-platform sandboxing (Linux bubblewrap, Windows managed sandbox, macOS seatbelt) and includes comprehensive telemetry, rollout tracing, and guardian approval pipelines. It is a fork of OpenAI's Codex CLI, with significant Rust rewrite and enhanced extensibility. While the codebase exhibits strong engineering practices (extensive testing, layered configuration, well-defined protocol types, and platform-agnostic abstractions), it is a large monorepo with many interdependent crates, some legacy modules, and platform-specific workarounds that reduce overall cohesion. Key reusable components include the protocol type definition crate (with clever no-op macro optimizations for build speed), a RAII OpenTelemetry timer, a RedactedString wrapper for secret handling, and token-aware text truncation utilities.
UmaDev is an AI-powered development agent that orchestrates multiple first-class AI coding CLIs (Claude Code, Codex, Grok Build, Kimi Code, OpenCode) as subprocesses, providing a unified terminal UI, governance, knowledge retrieval, and a spec-driven pipeline. It acts as a 'project manager' for AI coding agents, handling planning, execution, review, and acceptance. The architecture is a multi-crate Rust workspace with a TUI (ratatui), agent orchestration, governance engine, knowledge retrieval (BM25 + optional local embeddings), contract validation, and cross-platform process management. It is distributed via npm with platform-specific binaries. The codebase demonstrates high craftsmanship with pervasive fail-open patterns, bounded resource usage, and extensive testing. Key subsystems include a deterministic governance engine with 113 content checks, a hybrid retrieval system with CJK-aware tokenization, a cross-process run lock with boot-id liveness, and a host-owned Git commit transaction with rollback. While primarily an application, it includes a curated knowledge corpus (~400 markdown files) and an offline embedding model, giving it a hybrid character. The repository is actively maintained with a clear release pipeline and CI/CD. Notable design patterns include: fail-open governance (every check wrapped in catch_unwind), bounded file system operations (RootedDir with no-follow, size limits), circuit breaker for compaction, and source tree fingerprinting for freshness. The codebase is well-structured with clear separation of concerns across 11 crates, though some files are large hotspots (e.g., app.rs, director_loop.rs). Overall, it is a sophisticated, production-quality tool for AI-assisted software development.
This repository contains a minimal coding agent designed to work with local Ollama models. The agent operates in a loop, processing user requests by inspecting and modifying the workspace through a set of structured tools (file operations, shell commands, etc.). It implements six core components: live repo context collection, prompt shaping with cache reuse, structured tools with validation and permissions, context reduction and output management, transcript and memory persistence, and bounded delegation to subagents. The agent is implemented in a single Python file (`mini_coding_agent.py`) and is intended to be run via a CLI entry point. It requires an Ollama server running with a compatible model (default: qwen3.5:4b). The code is clean, well-structured, and demonstrates thoughtful agent design patterns, making it a useful reference for building similar agents.
This repository is a hybrid: it is primarily a curated awesome list for Claude Code resources, but it is backed by a real automation layer that treats the README as a generated artifact. The single source of truth is a CSV resource table plus a config.yaml that defines category ordering; a Python script (generate_readme.py) renders the README, ticker SVGs, and recently-added carousels. The repo also includes GitHub Actions workflows that validate new resource submissions via issue forms, handle maintainer approval/rejection commands, and automatically open PRs to add approved resources. This makes it less of a static list and more of a self-maintaining content pipeline.
DeepTide is a cross-platform terminal AI coding agent. The Rust workspace (deeptide-core, deeptide-host, deeptide-cli, deeptide-gui) implements the production agent engine: an interactive REPL session with an agent loop, tool dispatch (bash, edit, web, git, memory, MCP, LSP, vision, cron, planning), cross-provider model routing with local-first/cloud-escalation policies, cost tracking, tokens-per-second telemetry, tool usage observability, durable memory with BM25 retrieval ranking, session import/checkpoint/rewind, sensitive-file read protection, and a theme system for syntax highlighting and diffs. The CLI and GUI share a common host layer, and the npm package downloads prebuilt Rust binaries. Architecturally it is a multi-crate Rust workspace plus supporting native components: native/ds4 is a hard-forked DeepSeek V4 Flash Metal inference engine for local LLM execution, and native/dsgo is a Swift gateway that translates between Anthropic and OpenAI protocols with SSE streaming. Primary entry points are crates/deeptide-cli/src/main.rs (CLI), crates/deeptide-gui/src/main.rs (eframe/egui GUI), and bin.js (Bun launcher that delegates to @paean-ai/zero-cli). External dependencies include DeepSeek/Anthropic/OpenAI-compatible APIs, local Metal GPU inference, and GitHub Releases for binary distribution. The repo also contains 50+ self-contained HTML5 Canvas pixel-art games and Three.js showcases under samples/, which appear to be AI-generated demo content demonstrating agent output quality rather than core product code. These share heavily duplicated boilerplate (seeded RNG, localStorage save/load, bilingual i18n, screen state machines). The core Rust code is well-tested with integration tests covering routing, cost tracking, memory ranking, tool result summarization, and slash commands; live API tests are marked #[ignore]. Hybrid note: while primarily an application, it contains research-oriented elements such as memory capture benchmarks (benchmarks/auto_capture_bench.py), prompt cache layout benchmarks (benchmarks/cache_memory_bench.py), and deterministic puzzle generation algorithms with uniqueness verification.
Continue is an open-source AI coding agent that provides chat, autocomplete, edit, and agent capabilities across VS Code, JetBrains IDEs, and a standalone CLI. The architecture centers on a shared TypeScript core (`core/`) that handles LLM interactions, context providers, tool definitions, indexing, and configuration management. This core is consumed by three frontends: a VS Code extension (`extensions/vscode/`), a JetBrains plugin (`extensions/intellij/`), and a CLI (`extensions/cli/`) with a React-based TUI. A separate React GUI (`gui/`) runs as a webview in the IDE extensions. The system supports 50+ LLM providers through a plugin architecture in `core/llm/llms/` and uses a YAML-based configuration system (`packages/config-yaml/`) with a Hub for sharing assistants. Codebase indexing uses a content-addressed Merkle tree approach with a Rust sync library (`sync/`). The repository is now read-only and deprecated as of the final 2.0.0 release, with the team recommending users migrate to the CLI.