Wave of open-source tools for multi-agent coding systems launches on Hacker News
Developers are releasing frameworks for orchestrating teams of AI coding agents with persistent state, recovery, and hierarchical delegation.
What to know
- At least a dozen open-source projects for multi-agent coding systems landed on Hacker News in a single week (Sept 16–22), showing rapid developer interest in solving operational challenges of agent teams.
- Core problems being tackled: persistent task state recovery after interruptions, hierarchical delegation of work across agent teams, sandboxing and cost control, and real-time coordination of agent state and dependencies.
- Tools span multiple approaches: orchestration frameworks (Oh My Subagents, Fractal), architecture modeling for agents (Viaduct with MCP), local model support, workspace UIs, and sandbox environments—no single dominant pattern yet.
- Shift from ad-hoc agent spawning (polling loops, state loss on interruption) toward durable, accountable team structures with explicit task ownership, persistent checkpoints, and recovery guarantees.
ringlochid Creator of Oh My Subagentsigrlgkv Creator of ViaductDiogo Almeida AI architecture advisorsimone20a Creator of Orcristspuentesp Creator of Lain
How it unfolded 12 developments, newest first · click a bar or a number to jump posts
-
12
Fractal hierarchical agent loop framework released
An open-source tool for building hierarchical agent loops was posted, enabling deeper teams of agents to operate recursively without global polling.
-
I think this space is very untapped. Models are interesting, but I am absolutely obsessed with some things I've been researching/working on for the past few years:Fractal tool discovery: tool taxonomy where an agent can "drill deeper" to find what specific tool it's looking for. Helps if/when polluting context with a zillion (mostly unnecessary)…
2 more of the top 3 · 24 posts in this stretch
-
ringlochid/oh-my-subagents # devopsish https:// github.com/ringlochid/oh-my-su bagents
-
> Helps if/when polluting context with a zillion (mostly unnecessary) tools.Ugh, I can't help but respond to this one point. The fact that this is even an issue in the current year just tells us how screwed the software field is in a lot of ways. I don't mean that in existential terms, but of how divided we've become in terms of what's happened to…
-
-
11
DiscoBox AI agents in disposable sandboxes published
An open-source tool for running AI coding agents in isolated, disposable sandbox environments became available on GitHub.
-
The headline graph is kind of bizarre.For some reason they're comparing their harness running on Astra xhigh to Codex with Astra max?---Also worth noting that OpenAI just added support for async tool calling to their harness, which isn't 1:1 with this approach, but is slowly ramping up in being able to provide something similar.A big part of why…
1 more of the top 2 · 2 posts in this stretch
-
I think this is what I’ve been waiting for! Live interactions with models are fundamentally asynchronous! This will be great for interactivity.What a good time for harness design. Just while OpenCode is growing up a bit and focusing on their harness. I’m delighted to see people focusing on good general solid harness design principles.Someone make…
-
-
10
Orcrist desktop coding agent with fresh task harness released
A desktop coding agent that generates a custom harness for each task was posted to GitHub, offering a lightweight alternative to pre-configured agent systems.
-
9
Oh My Subagents framework for persistent agent teams released
A Python framework for durable multi-agent orchestration became available, enabling managers to delegate tasks to child agents with persistent state recovery across interruptions. The system commits assignments and checkpoints locally, allowing parent processes to continue even after terminal closure or provider session interruption.
“Ad-hoc delegation is easy to begin and surprisingly hard to operate. A parent spawns children, polls them, reconstructs ownership from chat, and hopes that a closed terminal or interrupted provider session did not erase the only useful account of what happened.”
— Oh My Subagents docs -
8
Project Terminal macOS workspace for agents announced
A macOS workspace application integrating terminals and coding agents was posted, offering a unified interface for agent interaction.
-
7
Viaduct C4 modeling tool for agents launches
A free C4 architecture modeling tool designed for AI agents to read and update documentation was released, allowing agents to interact with system architecture models via an MCP server. The tool works in the browser with no backend storage required and can import/export OpenAPI specs.
“An element exists once and appears on every level it belongs to, so a rename or a new dependency shows up everywhere at once.”
— Viaduct docs -
6
Motif shared memory system for agent teams released
An open-source shared memory framework designed for teams of AI coding agents became available, enabling better coordination across multiple agents.
-
5
Local-coder tool for teams of coding agents launches
A GitHub project enabling developers to build teams of coding agents using local models was posted to Hacker News.
-
4
Lain code graph and agent coordinator released
A structural code graph framework for coordinating multiple coding agents became available, addressing the challenge of maintaining state across agent interactions.
- 1 day quiet
-
3
Thoughts on typesafe coding agent architecture published
Diogo Almeida published design guidance on building typesafe coding agents, contributing to emerging best practices in the space.
- 1 day quiet
-
2
Sandboxing and VM costs for AI agents documented
A Velo Workspaces post analyzed the infrastructure costs of running coding agents in isolated VMs, addressing operational concerns as teams scale agent deployments.
-
1
Terminal-native parallel coding agent setup shared
A developer published a blog post describing a terminal-native approach to running multiple coding agents in parallel, signaling early interest in practical multi-agent orchestration.
What people are saying 19 voices from 1 site · best of 26 · verbatim
- Sep 23
-
This is so interesting.I'm unlocking savings (in both speed and costs) but using a different way of thinking by integrating Decision models:In short, LLMs spend so many turns on discovery, triage, and safety checks for everyday tasks. A decision model cuts those turns by preparing the relevant context before the prompt and handling the simple…
-
Crush [1] has had async tools for a long time, as has Claude Code.They work well, except that the agent will often simply call wait() immediately after, and also do it with a longer timeout, causing crashes/hangs to really slow down the process; whereas immediate tool calls have a built-in 30s timeout in most harnesses.I imagine this one will…
-
Is there a TUI or GUI I can use this with? Or is unreal-agent-runner intended to be called from an existing app? How does that work?
- Sep 22
-
I suppose GPT models throw shade when your input is less favorable to Codex:“On the surface, Unreal Agent achieves the same outcomes with fewer model turns and fewer input tokens.”(Emphasis mine)
-
Funny, I solved this problem by having Claude write a pi extensionhttps://github.com/Pyrolistical/pi-notifyNow my pi agent setups its own trigger to notify itself when a background process is done
-
Isn't that what codemode is for?mcp with 2 tools: execute(code) and search(query)all other tools, mcps, apis, whatever are encapsulated by the one interface. new tools don't bloat the agent's context, and it can write its own code to perform more advanced and batch operations against the available tools (executed within a sandbox).executor is a…
-
For everyone self-hosting models, optimising for cost is an anti-feature. It makes the results worse for no benefit (except a little speed).What I'd love to see is a harness that deeply optimises for the best results obtainable out of non-frontier models. Many of these have 1M context windows, and most of it remains unused and under utilised in…
-
OK, but isn't this just programmatic tool calling, which is beneficial in some cases but not others?This is available in DSH via PTC mode.Basically, instead of the completion API returning a sequence of tool calls, it returns a program that invokes the tools.I think most are returning Typescript programWith ordinary tool calling, you may have to…
-
not a fan of README's where there's no clear way to run the program, like a getting started or how to build. i think this is the opposite of a docu-monster where AI documents everything, but clearly more documentation is better than lack of.
-
So I commented on this in passing in a deeper thread (in re: the potential trademark issue - https://news.ycombinator.com/item?id=49807884), but I think this is serious enough for its own top-level thread.How likely is it that they get sued into the ground in a year? They might have a strong suite of offerings even as soon as six months from now…
-
The critique of CLI-oriented SDKs is fair. I'd want to see how it behaves on long-horizon tasks where the "no sub-agents" constraint starts to bite
-
i have found agents to be excellent at using CLIs, which are fractal-like. i built this reddit ads api cli and my agent immediately starting introspecting it in a "drill deeper" manner:
-
Have you had workflows/benchmarks to test this on? I'm primarily interested if there are real use cases that would benefit.
-
I've also explored s/Fractal tool discovery/Skill tree approaches, seems to work pretty well when you stick the equivalent of XREFs in the frontmatter.
-
Fractal tool discovery is a fascinating idea! Have you worked on implimenting this into any agent harnesses already to any success? My first impression is allowing the agent to fork itself, not unlike launching a subagent, then returning it's response back to the main agent.
-
Oh very nice, can you also compare it to https://maki.sh?Would be interesting to compare to a harness optimizing for cost reduction too.
-
In my projects i use unreal and me and my team have tested out multiple things. We have settled on no MCP complications whatsoever other than simply exposing the python scripting from the editor + a export step that can write the blueprints and asset data into plaintext so that the bot can grep them. This has given by far the strongest results…
-
If the frontier labs (well, I guess just Anthropic) would go full OAuth support even on a subscription we would see, an even bigger, explosion in harness improvements. I maintain that there is a ton of low-hanging fruit and new ideas/concepts that should be tried but the costs are holding people back (using API pricing only).It's both expensive to…
-
happy to see more harnesses in the Go space, i've had a fun time building https://github.com/ChaseRensberger/wingmani think the language primitives map well to the natural desire to use these tools across networks and in concurrent workloads.another Go harness i keep up to date with is: https://github.com/boldsoftware/shelley (exe.dev team)