RA.Aid

If I were to recommend one to someone today, I might pick RooCode. I'd suggest checking out boomerang mode and RooFlow on GitHub. Here are some others that I've tried and could recommend, in no particular order: - https://github.com/ai-christianson/RA.Aid - https://github.com/anthropics/claude-code - https://github.com/block/goose - https://github.com/hotovo/aider-desk I've also created a few "agents" to do specific tasks using Probe[0] as an MCP server, although I'm sure you could create a full-fledged agent with it if you wanted to. [0] https://github.com/buger/probe

In RA.Aid (https://github.com/ai-christianson/RA.Aid) we support normal tool calling, but we also have a our own agent backend called Code Is All You Need. The way it works is we prompt the LLM to generate code to call the tools directly. We then parse the AST to validate it and restrict it to a very small/safe subset (whitelisted function calls.) HF smolagents does something similar and there are a few papers out there validating the approach.

If you want one that was shipped 3 months earlier, is true FOSS (Apache 2.0,), and can even work reasonably with qwen-32b-coder-instruct, check https://github.com/ai-christianson/RA.Aid Disclaimer: am maintainer If you try it and have feedback, I'm very curious to hear it.

One problem we've had developing autonomous SWE agents (https://github.com/ai-christianson/RA.Aid) is that open models just haven't been performing near sonnet on controlling the agent. Our experience is echoed by many other agent devs out there, and you can see it for yourself if you try deepseek (v3 or r1) vs sonnet in any agentic product. Do you think that your training setup could help train these models to be better at agentic work?

Very cool! I'm working on a similar agent, but FOSS (https://github.com/ai-christianson/RA.Aid) --It'll be really interesting to see how the GitHub agent works. On first impressions, it looks like they are taking the route of integrating tightly with VSCode, which means they'll be competing with Cline, Cursor, and Windsurf. IMO it might be good for them to release this on the web, similar to the replit agent. Integration with GitHub directly would be awesome.

I'm in your boat with having to write a significant amount of English documents. I always write them myself, and have ChatGPT analyze them as well. I just had a thought - I wonder if I could paste in technical documentation, and code, to validate my documentation? Will have to try that later. CoPilot is used for simple boilerplate code, and also for the autocomplete. It's often a starting point for unit tests (but a thorough review is needed - you can't just accept it, I've seen it misinterpret code). I started experimenting with RA.Aid (https://github.com/ai-christianson/RA.Aid) after seeing a post on it here today. The multi-step actions are very promising. I'm about to try files-to-prompt (https://github.com/simonw/files-to-prompt) mentioned elsewhere in the thread. For now, LLMs are a level-up in tooling but not a replacement for deve

In their tool comparison section they did mention aider, but they did not mention RA.Aid (https://github.com/ai-christianson/RA.Aid, Apache 2.0), which puts aider in a full-fledged development agent. The experience of using RA.Aid is similar to using cline (or Devin), but it is a simple CLI command and not tied to any IDE. In fact you could use tmux, nvim, and RA.Aid in chat mode and have an experience that is not far off from vscode with cline, cursor, or windsurf.

If you want something like composer but want to stay fully FOSS/CLI oriented, check out https://github.com/ai-christianson/RA.Aid

Hello HN, I've recently been trying out and using the latest and greatest software development agents, including Windsurf and Cursor. Before using those, I had been using aider for everyday practical SWE tasks, but aider is not very agentic --it's best at one-off, scoped programming tasks. I did a quick test by putting aider in a ReAct loop using create_react_agent from langchain. To my surprise, it ended up working very well --it's able to solve programming tasks that Windsurf has failed at. The agent has access to many tools such as ripgrep, fuzzy find, filesystem navigation, etc. One tool, in particular, is an expert tool where it can use a dedicated reasoning model, such as OpenAI's o1, to reason about logic, bugs, or complex planning tasks. It's been really interesting to see when and how it decides to use these tools. The project do