I'm looking forward to trying this. I've had a positive but high-variance experience with Gastown[1], which is in the same genre. I hope that Scion does better.
My main complaints with Gastown are that (1) it's expensive, partly because (2) it refuses to use anything but Claude models, in spite of my configuration attempts, (3) I can't figure out how to back up or add a remote to its beads/dolt bug database, which makes me afraid to touch the installation, and (4) upgrading it often causes yak shaving and lost context. These might all be my own skill issues, but I do RTFM.
But wow, Gastown gets results. There's something magic about the dialogue and coordination between the mayor and the polecats that leads to an even better experience than Claude Code alone.
Really interesting to see Google's approach to this.
Recently I shared my approach, Optio, which is also an Agent Orchestration platform: https://news.ycombinator.com/item?id=47520220
I was much more focused on integrating with ticketing systems (Notion, Github Issues, Jira, Linear), and then having coding agents specifically work towards merging a PR.
Scion's support for long running agents and inter-container communication looks really interesting though. I think I'll have to go plan some features around that. Some of their concepts, make less sense to me, I chose to build on top of k8s whereas they seem to be trying to make something that recreates the control plane. Somewhat skeptical that the recreation and grove/hub are needed, but maybe they'll make more sense once I see them in action the first time.
Isolation over constraints sounds like the right philosophy. Containers give you a boundary but not vis into what ran inside them. Curious how much execution context Scion surfaces, w/o that you're still in a position similar to the LiteLLM attack where something can run and cause damage before you know it happened.
[primary author and architect of scion here]
There are several layers of state and telemetry - first is provided by the hook system available in most harnesses, then for those that provide OpenTelemetry -that is normalized and forwarded raw (preserving both) to a cloud collector. Finally - some activities are "self reported" by agents using a built-in toolset that can be reflected in the control plane
[primary author and architect of scion here] The missing features are mostly by design - this is closer to what the gastown plans as "gascity" - bring your own orchestration characters and definition.
> This project is early and experimental. Core concepts are settled, but expect rough edges. Local mode: relatively stable - Hub-based workflows: ~80% verified - Kubernetes runtime: early with known rough edges
i guess gastown is a better choice for now? idk i don't feel good about "relatively stable"
I want to experiment more with agents but my employer only pays for Claude Code, and TOS disallows using the subscription API for other purposes. Anyone else in the same boat? Token based pricing also gets expensive fast.
Agent orchestration is one side of the problem. The other side
is: where does the data go?
When agents process EU user data (names, emails, IBANs) and
route it to US model providers, that's a GDPR violation.
I open sourced a routing layer that detects PII in prompts and
forces EU-only inference when personal data is found:
https://github.com/mahadillahm4di-cyber/mh-gdpr-ai.eu
Their agent tooling is shaping up to be the well known issue of product cancellation. They have how many different takes on this now? (gemini-cli, antigravity, AI studio, this, Gemini app)
I've not been impressed with any of them. I do use their ADK in my custom agent stack for the core runtime. That one I think is good and has legs for longevity.
The main enterprise problem here is getting the various agent frameworks to play nice. How should one have shared runtimes, session clones, sandboxes, memory, etc between the tooling and/or employees?
Not if you go custom, you have unlimited latitude, examples...
I modified file_read/write/edit to put the contents in the system prompt. This saves context space, i.e. when it rereads a file after failed edit, even though it has the most recent contents. It also does not need to infer modified content from read+edits. It still sees the edits as messages, but the current actual contents are always there.
My AGENTS.md loader. The agent does not decide, it's deterministic based on what other files/dirs it has interacted with. It can still ask to read them, but it rarely does this now.
I've also backed the agents environment or sandbox with Dagger, which brings a number of capabilities like being able to drop into a shell in the same environment, make changes, and have those propagate back to the session. Time travel, clone/fork, and a VS Code virtual FS are some others. I can go into a shell at any point in the session history. If my agent deletes a file it shouldn't, I can undo it with the click of a button.
I can also interact with the same session, at the same time, from VS Code, the TUI, or the API. Different modalities are ideal for different tasks (e.g. VS Code multi-diff for code review / edits; TUI for session management / cleanup).
jj will not achieve meaningful adoption until git interop is improved and there is a big enough win to change a core work tool. Lack of git-lfs is a blocker where I work and asking all the devs to change their git habits for a shop that doesn't use rebase (as I understand the main issue jj aims to make better)... the ROI just doesn't appear to be there.
I swore to not be burned by google ever again after TensorFlow. This looks cool, and I will give this to my Codex to chew on and explain if it fits (or could fit what I am building right now -- the msx.dev) and then move on. I don't trust Google with maintaining the tools I rely on.
I'm looking forward to trying this. I've had a positive but high-variance experience with Gastown[1], which is in the same genre. I hope that Scion does better.
My main complaints with Gastown are that (1) it's expensive, partly because (2) it refuses to use anything but Claude models, in spite of my configuration attempts, (3) I can't figure out how to back up or add a remote to its beads/dolt bug database, which makes me afraid to touch the installation, and (4) upgrading it often causes yak shaving and lost context. These might all be my own skill issues, but I do RTFM.
But wow, Gastown gets results. There's something magic about the dialogue and coordination between the mayor and the polecats that leads to an even better experience than Claude Code alone.
1. https://github.com/gastownhall/gastown/
Really interesting to see Google's approach to this. Recently I shared my approach, Optio, which is also an Agent Orchestration platform: https://news.ycombinator.com/item?id=47520220
I was much more focused on integrating with ticketing systems (Notion, Github Issues, Jira, Linear), and then having coding agents specifically work towards merging a PR. Scion's support for long running agents and inter-container communication looks really interesting though. I think I'll have to go plan some features around that. Some of their concepts, make less sense to me, I chose to build on top of k8s whereas they seem to be trying to make something that recreates the control plane. Somewhat skeptical that the recreation and grove/hub are needed, but maybe they'll make more sense once I see them in action the first time.
Isolation over constraints sounds like the right philosophy. Containers give you a boundary but not vis into what ran inside them. Curious how much execution context Scion surfaces, w/o that you're still in a position similar to the LiteLLM attack where something can run and cause damage before you know it happened.
[primary author and architect of scion here] There are several layers of state and telemetry - first is provided by the hook system available in most harnesses, then for those that provide OpenTelemetry -that is normalized and forwarded raw (preserving both) to a cloud collector. Finally - some activities are "self reported" by agents using a built-in toolset that can be reflected in the control plane
They kinda buried the code deep in their docs:
https://github.com/GoogleCloudPlatform/scion
Exactly, I actually starred this in late March and hadn't made my way back to it yet. Glad somebody posted, looks very interesting.
This seems to be in the direction of Gas Town but missing some of the core features. Having formulas has been game changing.
[primary author and architect of scion here] The missing features are mostly by design - this is closer to what the gastown plans as "gascity" - bring your own orchestration characters and definition.
If you look at this orchestration example
https://github.com/ptone/scion-athenaeum
its just markdown - Scion is the game engine
(a port of gastown to run on scion is in progress)
> This project is early and experimental. Core concepts are settled, but expect rough edges. Local mode: relatively stable - Hub-based workflows: ~80% verified - Kubernetes runtime: early with known rough edges
i guess gastown is a better choice for now? idk i don't feel good about "relatively stable"
imagine thinking gas town is a better choice over _literally anything else_
Reading this headline, I rather thought of a different SCION:
> https://en.wikipedia.org/wiki/SCION_(Internet_architecture)
I want to experiment more with agents but my employer only pays for Claude Code, and TOS disallows using the subscription API for other purposes. Anyone else in the same boat? Token based pricing also gets expensive fast.
This runs stock Claude Code in containers, should be completely fine for TOS
this is very cool! i recently hacked on something similar https://github.com/s2-streamstore/parallax
and also wrote about it https://s2.dev/blog/distributed-ai-agents
Agent orchestration is one side of the problem. The other side is: where does the data go?
My brain keeps wanting to pronounce it Tomb Raider style, like /ˈskiː ɒn/.
Their agent tooling is shaping up to be the well known issue of product cancellation. They have how many different takes on this now? (gemini-cli, antigravity, AI studio, this, Gemini app)
I've not been impressed with any of them. I do use their ADK in my custom agent stack for the core runtime. That one I think is good and has legs for longevity.
The main enterprise problem here is getting the various agent frameworks to play nice. How should one have shared runtimes, session clones, sandboxes, memory, etc between the tooling and/or employees?
It's all just system prompts under the hood and nothing more.
[primary author and architect of scion here] Actually - there are two other big parts: a CLI and a control plane
Not if you go custom, you have unlimited latitude, examples...
I modified file_read/write/edit to put the contents in the system prompt. This saves context space, i.e. when it rereads a file after failed edit, even though it has the most recent contents. It also does not need to infer modified content from read+edits. It still sees the edits as messages, but the current actual contents are always there.
My AGENTS.md loader. The agent does not decide, it's deterministic based on what other files/dirs it has interacted with. It can still ask to read them, but it rarely does this now.
I've also backed the agents environment or sandbox with Dagger, which brings a number of capabilities like being able to drop into a shell in the same environment, make changes, and have those propagate back to the session. Time travel, clone/fork, and a VS Code virtual FS are some others. I can go into a shell at any point in the session history. If my agent deletes a file it shouldn't, I can undo it with the click of a button.
I can also interact with the same session, at the same time, from VS Code, the TUI, or the API. Different modalities are ideal for different tasks (e.g. VS Code multi-diff for code review / edits; TUI for session management / cleanup).
Don't forget a while loop and a TODO.md
Disapointing google of all places uses git worktrees instead of jj workspaces.
jj will not achieve meaningful adoption until git interop is improved and there is a big enough win to change a core work tool. Lack of git-lfs is a blocker where I work and asking all the devs to change their git habits for a shop that doesn't use rebase (as I understand the main issue jj aims to make better)... the ROI just doesn't appear to be there.
I swore to not be burned by google ever again after TensorFlow. This looks cool, and I will give this to my Codex to chew on and explain if it fits (or could fit what I am building right now -- the msx.dev) and then move on. I don't trust Google with maintaining the tools I rely on.
nice plug