hardsnow

4 ms·

1.
by hardsnow 2mo ago
GitHub doesn’t exactly make it easy to configure agent access securely. In fact, their regular access tokens and app credentials don’t provide granular enough controls to give direct access to private repos securely. Even if tokens are tigh
2.
by hardsnow 5mo ago
I’ve been using email as an async channel with agents. Email does proper long-form async and native threaded communication extremely well and IMO is the best match UX-wise. The system I’ve developed for this is open source and detailed at
3.
by hardsnow 5mo ago
I seriously think this great! I’ve been saying that email is the right interface for agents for a while now. It is available anywhere, natively threaded, and works for asynchronous long-form communication. Comes with great clients as well.
4.
by hardsnow 5mo ago
This is a sandbox escape pentest so the only tooling needed is Claude Code and a simple prompt that asks it to follow a workflow: https://github.com/airutorg/airut/blob/main/workflows/sandbo...
5.
by hardsnow 5mo ago
I’ve recently set up nightly automated pentest for my open-source project. I’m considering starting to publish these reports as proof of security posture. If the cost of security audit becomes marginal, it would seem reasonable to expect pr
6.
by hardsnow 5mo ago
I’ve been developing an open-source version of something similar[1] and used it quite extensively (well over 1k PRs)[2]. I’m definitely believer of the “prompt to PR model”. Very liberating to not have to think about managing the agent sess
7.
by hardsnow 5mo ago
I would agree with this. I recently tried to figure out how to properly secure agent-authored code in GitHub Actions. I believe I succeeded in doing this[1] but the secure configuration ended up being so delicate that I don’t have high hope
8.
Open-source system that runs Claude Code tasks from email and Slack(airut.org)
2 points by hardsnow 6mo ago|0 comments
9.
by hardsnow 6mo ago
OpenShell is the gem here indeed. A lot of good ideas like network sandbox that does TLS decryption and use of policy engine to set the rules. However: > Credentials never leak into the sandbox filesystem; they are injected as environmen
10.
by hardsnow 6mo ago
I’ve been running this with workloads accessing Anthropic, GitHub, Gemini, and AWS & CF R2 APIs for a while now, and have not ran into issues. I’m sure there’s an API out there that won’t work out of the box but I’m positive that suppor
11.
by hardsnow 6mo ago
This is the right approach. I built a similar system to https://github.com/airutorg/airut - couple of learnings to share: 1) Not all systems respect HTTP_PROXY. Node in particular is very uncooperative in this regard.
12.
Securing Agentic AI Is a Probabilistic Problem(haulos.com)
2 points by hardsnow 6mo ago|0 comments
13.
From one-shot to agentic diagnostic analysis(haulos.com)
1 points by hardsnow 6mo ago|0 comments
14.
by hardsnow 7mo ago
Yep - requires the client to trust the SSL cert of the proxy. Cooperative clients that support eg HTTP_PROXY may be easier to support, but for Airut I went for full transparent mitmproxy. All DNS A requests resolve to the proxy IP and proxy
15.
by hardsnow 7mo ago
Alternative, and more robust approach is to give the agent surrogate credentials and replace them on the way out in a proxy. If proxy runs in an environment to which agent has no access to, the real secrets are not available to it directly;
16.
Show HN: Airut – Sandboxed Claude Code over Email and Slack(github.com)
1 points by hardsnow 7mo ago|0 comments
17.
by hardsnow 7mo ago
Container isolation is a good foundation, but one layer worth adding is network sandboxing. A filesystem-sandboxed agent can still exfiltrate data over the network if it gets prompt-injected — domain allowlists and egress filtering can redu
18.
Show HN: Airut – Sandboxed Claude Code sessions over email(github.com)
2 points by hardsnow 7mo ago|0 comments
19.
by hardsnow 4y ago
As someone with first-hand experience in implementing pass-through AR, I found the review well-made (e.g., they measured photon-to-photon latency) and extremely interesting. Here's a few observations I made: 1. 30 Hz frame rate and 50m