Google open-sources experimental agent orchestration testbed Scion
by timbilt on 4/7/2026, 1:39:15 PM
<a href="https://googlecloudplatform.github.io/scion/overview/" rel="nofollow">https://googlecloudplatform.github.io/scion/overview/</a>
https://www.infoq.com/news/2026/04/google-agent-testbed-scion/
Comments
by: hackerman70000
Six months from now half of these abstractions will have been renamed or removed once real users push back on the cognitive overhead. Google has a pattern of releasing infrastructure that's perfectly shaped for Googles problems and awkward for everyone else's
4/7/2026, 4:26:40 PM
by: sowbug
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.<p>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.<p>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.<p>1. <a href="https://github.com/gastownhall/gastown/" rel="nofollow">https://github.com/gastownhall/gastown/</a>
4/7/2026, 5:41:40 PM
by: w10-1
The secondary meaning of "scion" they don't cite is "a descendant of a notable family"<p>The unfortunate meaning: this is who will rule you, so get used to it?<p>More generally: they take a loose, natural-language approach to orchestration which would be more promising than scalable.<p>In design, I try to reduce all unnecessary degrees of freedom, and focus effort on the necessary ones. Otherwise, there's too much play/slop as the probabilities fan out. It translates to fast starts but huge blockers trying to sort out combinatorics - classic turtle and hare.<p>So, no.
4/7/2026, 6:22:07 PM
by: tornikeo
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.
4/7/2026, 5:09:00 PM
by: armanj
> 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<p>i guess gastown is a better choice for now? idk i don't feel good about "relatively stable"
4/7/2026, 5:57:21 PM
by: simple10
They kinda buried the code deep in their docs:<p><a href="https://github.com/GoogleCloudPlatform/scion" rel="nofollow">https://github.com/GoogleCloudPlatform/scion</a>
4/7/2026, 4:47:45 PM
by: kvanbeek
This seems to be in the direction of Gas Town but missing some of the core features. Having formulas has been game changing.
4/7/2026, 5:58:37 PM
by: aleph_minus_one
Reading this headline, I rather thought of a different SCION:<p>> <a href="https://en.wikipedia.org/wiki/SCION_(Internet_architecture)" rel="nofollow">https://en.wikipedia.org/wiki/SCION_(Internet_architecture)</a>
4/7/2026, 5:38:11 PM
by: cedws
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.
4/7/2026, 5:25:17 PM
by: meidad_g
[dead]
4/7/2026, 6:05:25 PM
by: Sattyamjjain
[dead]
4/7/2026, 5:58:43 PM
by: ninjahawk1
[dead]
4/7/2026, 6:15:03 PM
by: kumardeepanshu
[dead]
4/7/2026, 5:33:52 PM
by: aplomb1026
[dead]
4/7/2026, 5:31:10 PM
by: verdverm
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)<p>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.<p>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?
4/7/2026, 4:26:50 PM