2025-05-13 13:23:46

danvoronov · danvoronov · commit f9cfdfef47bc · 2025-05-13T13:23:49.000+03:00
diff --git a/Tools.md b/Tools.md
@@ -27,7 +27,7 @@ Closed or unmaintained projects: [aide](https://github.com/codestoryai/aide) (Fe
 | [Junie](https://www.jetbrains.com/junie/)                                                                                 | Junie is coding agent by JetBrains. Currently available in IntelliJ IDEA Ultimate and PyCharm Professional, with WebStorm soon to be added to this list.                                                                                                    |
 | ![icon](https://www.google.com/s2/favicons?domain=augmentcode.com&sz=24) [augment code](https://www.augmentcode.com/)     | The first AI coding assistant built for professional software engineers and large codebases.                                                                                                                                                                |
 | ![icon](https://www.google.com/s2/favicons?domain=continue.dev&sz=24) [Continue.dev](https://www.continue.dev/)           | An open-source AI coding assistant for IDEs that helps accelerate software development. It offers features like code generation, refactoring, and debugging assistance.                                                                                     |
-| ![icon](https://www.google.com/s2/favicons?domain=supermaven.com&sz=24) [Supermaven](https://supermaven.com/)             | Supermaven is VS Code, JetBrains IDEs, and Neovim extension that lets you write code 2x faster with AI.                                                                                                                                                     |
+| ![icon](https://www.google.com/s2/favicons?domain=supermaven.com&sz=24) [Supermaven](https://supermaven.com/)             | Supermaven is VS Code, JetBrains IDEs, and Neovim extension that lets you write code 2x faster with AI. On Nov 12, 2024, the team joined Cursor and is now working on Cursor Tab.                                                                           |
 | [MarsCode](https://www.marscode.com/)                                                                                     | Built-in AI programming assistant with capabilities like code completion, explanation, and debugging for faster development.                                                                                                                                |
 | ![icon](https://www.google.com/s2/favicons?domain=kilocode.ai&sz=24) [Kilo Code](https://kilocode.ai/)                    | Kilo Code is an open source AI agent VS Code extension. It helps you write code more efficiently by generating code, automating tasks, and providing suggestions.                                                                                           |
 | ![icon](https://www.google.com/s2/favicons?domain=twinny.dev&sz=24) [Twinny](https://twinny.dev/)                         | The most no-nonsense, locally or API-hosted AI code completion plugin for Visual Studio Code - like GitHub Copilot but 100% free and private.                                                                                                               |
@@ -40,7 +40,7 @@ Closed or unmaintained projects: [aide](https://github.com/codestoryai/aide) (Fe
 | ![icon](https://www.google.com/s2/favicons?domain=tabnine.com&sz=24) [Tabnine](https://www.tabnine.com/)                  | Tabnine is the world’s most contextually aware AI software development platform, helping mature engineering teams speed up and simplify their entire development process. Context-aware suggestions based on your code and patterns.                        |
 | ![icon](https://www.google.com/s2/favicons?domain=gitlab.com&sz=24) [GitLab Duo](https://about.gitlab.com/gitlab-duo/)    | From planning and coding to securing and deploying, Duo is the only AI solution that supports developers at every stage of their workflow.                                                                                                                  |
 | [Amazon Q Developer](https://aws.amazon.com/q/developer/)                                                                 | Amazon Q Developer is plugin that provides you with code recommendations in real time and is an expert on AWS and is in the AWS Management Console.                                                                                                         |
-| [Gemini Code Assist](https://codeassist.google/)                                                                          | Google Gemini Code Assist brings the power of Gemini 2.0 to your IDE, with a chat interface that is aware of your code, automatic code completions as you write, and generation and transformation of full functions or files on demand.                    |
+| [Gemini Code Assist](https://codeassist.google/)                                                                          | Google Gemini Code Assist brings the power of Gemini to your IDE, with a chat interface that is aware of your code, automatic code completions as you write, and generation and transformation of full functions or files on demand.                        |
 
 Closed or unmaintained projects: [rubberduck](https://github.com/rubberduck-ai/rubberduck-vscode) (Feb 4, 2024).
 
@@ -52,7 +52,7 @@ Closed or unmaintained projects: [rubberduck](https://github.com/rubberduck-ai/r
 | ![icon](https://www.google.com/s2/favicons?domain=aider.chat&sz=24) [aider chat](https://aider.chat/)                                                                             | Coding agent CLI tool that allows you to interact with an AI directly from your terminal. It helps with code generation, modification, and understanding through a chat interface.                                                                                                       |
 | [Claude Code](https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/overview)                                                                                           | Coding agent Claude Code (Research Preview) is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster through natural language commands.                                                                                               |
 | [OpenAI Codex](https://github.com/openai/codex)                                                                                                                                   | Lightweight open-source coding agent from OpenAI that runs in terminal. Zero setup — just OpenAI API key and it  works. Full auto-approval, while safe + secure by running network-disabled and directory-sandboxed. Multimodal — pass in screenshots or diagrams to implement features. |
-| [Kwaak](https://github.com/bosun-ai/kwaak)                                                                                                                                        | Coding agent Kwaak enables you to run a team of autonomous AI agents right from your terminal, in parallel. You interact with Kwaak in a chat-like terminal interface.                                                                                                                   |
+| [Kwaak](https://github.com/bosun-ai/kwaak)                                                                                                                                        | Kwaak can be considered alpha software. Coding agent Kwaak enables you to run a team of autonomous AI agents right from your terminal, in parallel, with MCP support. You interact with Kwaak in a chat-like terminal interface.                                                         |
 | [Sidekick](https://github.com/geekforbrains/sidekick-cli)                                                                                                                         | Sidekick is an agentic CLI-based AI tool inspired by Claude Code, Copilot, Windsurf and Cursor. It's meant to be an open source alternative to these tools, providing a similar experience but with the flexibility of using different LLM providers while keeping the agentic workflow. |
 | [RA.Aid](https://www.ra-aid.ai/)                                                                                                                                                  | A standalone coding agent built on LangGraph's agent-based task execution framework. The tool provides an intelligent assistant that can help with research, planning, and implementation of multi-step development tasks.                                                               |
 | ![icon](https://www.google.com/s2/favicons?domain=plandex.ai&sz=24) [Plandex](https://plandex.ai/)                                                                                | Plandex is a top-tier coding agent with fully autonomous capabilities. By default, it combines the strengths of three top foundation model providers—Anthropic, OpenAI, and Google. Two subscription options: an Integrated Models mode cost $45 per month.                              |
@@ -98,6 +98,7 @@ Closed or unmaintained projects: [rubberduck](https://github.com/rubberduck-ai/r
 | **RAG for code**         | [Pieces](https://pieces.app/)                                                                                              | Chat with a developer [copilot](https://pieces.app/features/copilot) powered by the LLM of your choice. Recall all the activities on your device using the Pieces [Long-Term Memory](https://pieces.app/features/copilot/long-term-memory) & bring contextual awareness to your chats by adding any files or folders of code, websites, or saved code snippets. |
 | **Code security**        | ![icon](https://www.google.com/s2/favicons?domain=snyk.io&sz=24) [Snyk](https://snyk.io/platform/deepcode-ai/)             | DeepCode AI utilizing multiple fine-tuned AI models and security-specific data curated by top security specialists, to give you all the power of AI in your security, without any of the drawbacks.                                                                                                                                                             |
 |                          | [Strix](https://usestrix.com/)                                                                                             | Strix is a collaborative AI agent, spotting security issues before they become problems Strix lets you pentest using instructions. Craft intelligent, context-aware payloads, commands, wordlists, and scripts with a simple prompt.                                                                                                                            |
+|                          | [Sourcebot](https://docs.sourcebot.dev/docs/agents/review-agent)                                                           | The AI Code Review Agent is open source and packaged in Sourcebot.  This agent provides codebase-aware reviews for your PRs. For each diff, this agent fetches relevant context from Sourcebot and feeds it into an LLM for a detailed review of your changes.                                                                                                  |
 |                          | [Diamond](https://diamond.graphite.dev/)                                                                                   | Graphite Reviewer is now Diamond. Get immediate, actionable feedback on every pull request with Graphite’s codebase-aware AI.                                                                                                                                                                                                                                   |
 |                          | [Sonar](https://www.sonarsource.com/solutions/ai/ai-code-assurance/)                                                       | Ensure the quality and security of every line of AI generated code by instilling confidence using our code assurance workflow. AI coding assistants like GitHub Copilot, Amazon Q, and Google Code Assist, when combined with Sonar, can boost developers' productivity and ensure secure, stable code.                                                         |
 | **Files to Prompt**      | ![icon](https://www.google.com/s2/favicons?domain=repoprompt.com&sz=24) [Repo Prompt](https://repoprompt.com/)             | Repo Prompt structure your prompts and apply AI changes for you.                                                                                                                                                                                                                                                                                                |
diff --git a/eng_2025/05/2025-05-13-12-51.md b/eng_2025/05/2025-05-13-12-51.md
@@ -0,0 +1,28 @@
+It seems that using Claude code, Cursor, and others has become largely repetitive. The workflow usually involves planning the task (roadmap file), then giving commands to the agent to implement the plan in code.
+
+Thus, **task orchestration** is the next necessary thing for every agentic AI solution.
+
+I have already mentioned https://www.task-master.dev/, which is currently a popular solution due to MCP.
+
+---
+
+**aider**
+https://aider.chat/docs/scripting.html
+aider natively allows using simple **scripting** from the terminal to perform repetitive actions. There is also an additional Python API function for scripting, but it is not officially supported or documented.
+
+**Roo Code** | Boomerang Orchestrator (since ver 3.14.3)
+https://docs.roocode.com/features/boomerang-tasks
+They added "`🪃 Orchestrator`" as a built-in mode. It allows breaking down complex projects into smaller, manageable parts. Each sub-task is then executed in its own context, often using a different mode tailored for that specific task.
+
+---
+
+**Code Claude Code**
+https://github.com/RVCA212/codesys
+A project developing a Python SDK for **interacting with the Claude CLI** tool. The most effective way to use it is by mimicking your actual workflow. Supports resuming specific conversations by ID.
+
+**Cloud Code SDK**
+https://cloudcoding.ai/
+A programmable AI Coder SDK in Python - both locally and in a Sandbox cloud. You can think of it as a way to interact with Cursor or Claude code, at a low level with great control. But instead of using these applications, the project uses *its own agent* that can modify code and use its own built-in tools. Currently supports only OpenAI and Anthropic models. Works with or without Git repositories.
+
+
+#aider #claudecode #agentmode
diff --git a/ukr_2025/05/2025-05-13-12-51.md b/ukr_2025/05/2025-05-13-12-51.md
@@ -0,0 +1,30 @@
+Здається, що використання Claude code, Cursor та інших стало здебільшого повторюваним. Робочий процес зазвичай виглядає так: планування завдання (файл roadmap), потім команди агенту реалізувати план у код.
+
+Тим самим оркестрація завдань є наступною потрібною річчю для кожного агентного AI-рішення.
+
+Про https://www.task-master.dev/ я вже згадував, це зараз популярне рішення через MCP. 
+
+---
+
+**aider**
+https://aider.chat/docs/scripting.html
+aider нативно дозволяє використовувати простий **скриптинг** із терміналу, щоб виконувати повторювані дії. Також є додаткова функція Python API для скриптингу, але офіційно не підтримується та не документована.
+
+
+**Roo Code** | Boomerang Orchestrator (з ver 3.14.3)
+https://docs.roocode.com/features/boomerang-tasks
+Додали "`🪃 Orchestrator`" як вбудований режим. Дозволяє розбити складні проєкти на менші, керовані частини. Потім кожне підзавдання виконується у власному контексті, часто використовуючи інший режим, адаптований для цього конкретного завдання.
+
+---
+
+**Code Claude Code**
+https://github.com/RVCA212/codesys
+Проект, який розробляє Python SDK для **взаємодії з Claude CLI** tool. Найефективніший спосіб використання полягає в імітації вашого фактичного робочого процесу. Підтримка відновлення конкретних розмов за ID.
+
+
+**Cloud Code SDK**
+https://cloudcoding.ai/
+Програмоване AI Coder SDK на Python - як локально, так й у Sandbox  хмарі. Можна уявити це як спосіб взаємодіяти з Cursor або Claude code, на низькому рівні з великим контролем. Але замість використання цих застосунків, проект використовує власного агента, який може змінювати код та використовувати власні вбудовані інструменти. Наразі підтримує лише моделі OpenAI та Anthropic. Працює з або без Git-репозиторіїв.
+
+
+#aider #claudecode #agentmode 
