
CEO & Founder @NOFireAI | Fixing Reliability with AI | Author of ArgoCD in practice
Sparti, Greece
Joined October 2010
- Tweets1.3K
- Following1.6K
- Followers681
- Likes1.1K
Take your seat here: luma.com/w029h353
brig v0.4.0 gives new sandboxes their own networks by default on Linux and on macOS.
Run coding agents for different projects on one machine, while keeping their dev servers and test APIs separate.
In this demo, one sandbox tries to reach a web server in the other. The connection times out.
We then start both with --network shared, and the same request works.
You can run unrelated tasks side by side and enable network sharing when agents need to work together.
#OpenSource #Sandboxing
brig v0.4.0 limits where your coding agent can send your code and credentials.
This release brings egress policy enforcement on Linux! If brig cannot enforce the policy, then the run is refused.
Allow the services you need for the job. Then brig enforces your network policy outside the sandbox, so the rules also apply to the scripts and dependencies the agent runs. In this demo, an example domain works. Requests to Github API and a direct connection to 1.1.1.1 are blocked, even from a root shell inside the sandbox.
#OpenSource #Sandboxing
In San Francisco this Friday for #sftechweek?
Join our CTO, Panagiotis Moustafellos and DevRel Artemis Leonardou, for a private roundtable on AI, security and production reliability.
October 9 · 8–11 PM PT · San Francisco
#SFTechWeek #SRE
Spiros Economakis retweeted
Links:
Naive-N0.5-Flash — naive.ai/en/research/
OpenScience — github.com/synthetic-science…
Brig — github.com/brig-sh/brig
New engineers are shipping with #AI. Is your on-call training ready?
@artemisllm speaks at @sreday_com San Francisco on Oct 2, 11:30 a.m. PT.
Come meet her at the Harness office. Use SRE50 for 50% off tickets.
brig v0.3.0 is out.
Rootless Linux support, simpler setup and local app previews.
Thanks to @anan0s , @alexsapran, Maria Gkoutha, @pmoust and Vangelis Katsikaros.
Thanks also @p80n_sec of @EndorLabs for reporting the security issue.
What does “junior developer” mean when you can ship with AI from day one?
@artemisllm is speaking at @devconf_us:
“The End of the Junior Developer”
Sept 25 · 2:20 PM ET · Boston
We had a blast at Signals Conference! Thank you to the team behind all the hard work that went into the first #Reliability in the Age of AI conference in Europe.
@pmoust presented the principles behind our work and introduced #Brig, a free and open-source #sandbox for coding agents. It's part of our vision and how we're building the first platform for controlled autonomy in production operations in the age of AI, with safety guarantees.
Learn more about why we built our own microVMM in the comments.
Bookmark this if you run Claude Code / Codex with auto-approve.
Your agent can rm, curl, and leak keys. Sandbox map:
Local microVM (real VM on your machine)
• Docker Sandboxes - sbx run claude
• shuru - ephemeral, offline by default
• Brood Box - microVM + review diffs before they land
• Brig - open-source microVM, curated agent profiles
# curl -fsSL brig.sh/install | sh
Cloud (agent runs elsewhere)
• E2B - Firecracker, SDK-first (most used for products)
• Daytona - persistent agent workspaces
• Vercel Sandbox / Modal - hosted code exec
Weaker (shared kernel)
• DevContainers / plain Docker - fine for deps, not a blast-radius wall
Pick by job:
laptop agent → Docker SBX / shuru / Brig
Product feature → E2B
Spiros Economakis retweeted
Jev (RLCD) opens exciting opportunities. A general purpose classifier is what many usecases are after. LLMs are the wrong or expensive choice for many of the problems they are used for today.
Top open Qs? Training data set, what is it? shape/origin? How's the model eval'ed?
We invested in microVM isolation #sandboxing for @NOFireAI ’s work on controlled autonomy in production.
Today we’re open-sourcing that work as Brig, for anyone running AI coding agents on Mac or Linux.
Every component under Apache 2.0 license and free.
Get started with: `brig run claude`
Check it out at: brig.sh/
Spiros Economakis retweeted
While working on Ravel we noticed arrayef getting yanked, see: github.com/NOFireAI/ravel/co…, decided to keep the 0.3.9 pin. Now assessing.
This is why we develop in our #sandbox solution.
We are soon announcing the FOSS release of our sandboxing solution to the public.
Stay tuned.
Spiros Economakis retweeted
SELECT * FROM logs - but watch where it actually goes.
Turned #Ravel's codebase into a city you can explore. Every crate's a block and its height is LoC.
Press play and see: push the time range and blooms down -> decode RLOG blocks -> stream batches, etc.
github.com/NOFireAI/ravel
What if S3-compatible object storage was enough for most of your #observability / #SIEM stack?
We’re testing that with Ravel: logs, metrics, traces, PromQL, SQL + distributed search.
Still early, not prod-ready. Try to break it:
github.com/NOFireAI/ravel
Hugging Face could reconstruct 17,000+ attacker actions because they own the infrastructure the agent ran on, end to end. Most teams running agents today don't have that. linkedin.com/pulse/agent-wro…
For most of us, the only account of what an agent did is a log the agent itself produced, from inside the same environment it was operating in. If that agent is compromised, misconfigured, or just wrong, there's nothing independent to check it against. A log the agent wrote about itself is not evidence. It's the agent's word.
















