Islo deploys and runs autonomous coding agents in secure environments connected to your tools, so they can finish real engineering work.https://islo.dev/en-ushttps://islo.dev/blog/benchmarks-as-an-attack-vector/https://islo.dev/blog/benchmarks-as-an-attack-vector/How a Harbor regrade vulnerability let task-authored code read a host file, and what it means for enterprise evaluations and RLVR training.Thu, 22 Oct 2026 00:00:00 GMThttps://islo.dev/blog/verifier-tampering/https://islo.dev/blog/verifier-tampering/In a controlled Harbor run, the tests failed and a fairness judge caught the cheat, yet submitted code overwrote the final score. We examine the trust boundary inside verification and the Terminal-Bench community's work to secure it.Thu, 15 Oct 2026 00:00:00 GMThttps://islo.dev/blog/your-agent-evals-are-wrong/https://islo.dev/blog/your-agent-evals-are-wrong/Replaying real engineering work sounds straightforward until the agent needs the internet. The eval has to freeze that internet at the moment the work was done.Mon, 28 Sep 2026 00:00:00 GMThttps://islo.dev/blog/how-arimlabs-runs-agent-evaluations-at-scale-on-islo/https://islo.dev/blog/how-arimlabs-runs-agent-evaluations-at-scale-on-islo/ARIMLABS moved agent evaluations from local machines to an always-on Islo workflow with isolated computers, parallel trials, and inspectable reruns.Tue, 15 Sep 2026 00:00:00 GMThttps://islo.dev/blog/ai-coding-has-outgrown-the-laptop/https://islo.dev/blog/ai-coding-has-outgrown-the-laptop/AI coding began on developer laptops. Autonomous work needs shared, persistent, observable and governable cloud environments.Mon, 14 Sep 2026 00:00:00 GMThttps://islo.dev/blog/what-is-a-software-factory-5-products-to-evaluate-in-2026/https://islo.dev/blog/what-is-a-software-factory-5-products-to-evaluate-in-2026/What is a software factory? Compare Devin, Cursor, Linear, Factory, and Islo across autonomy, infrastructure, verification, governance, and fit.Mon, 17 Aug 2026 00:00:00 GMThttps://islo.dev/blog/introducing-islo-byoc/https://islo.dev/blog/introducing-islo-byoc/Run Islo coding agents on infrastructure you own—your AWS account or on-prem Linux machines—while Islo keeps operating the product for you.Thu, 30 Jul 2026 00:00:00 GMThttps://islo.dev/blog/how-alien-uses-ai-sims-to-continuously-test-its-deployment-platform/https://islo.dev/blog/how-alien-uses-ai-sims-to-continuously-test-its-deployment-platform/Alien runs AI agents called Sims on Islo to test deployment scenarios across clouds, architectures, and customer-controlled environments.Wed, 29 Jul 2026 00:00:00 GMThttps://islo.dev/blog/openai-could-have-stopped-hugging-face-attack/https://islo.dev/blog/openai-could-have-stopped-hugging-face-attack/How OpenAI's model escaped its evaluation sandbox through an allowlisted package proxy, and why host-level allow and deny is the wrong granularity for containing AI agents.Sun, 26 Jul 2026 00:00:00 GMThttps://islo.dev/blog/how-obert-built-a-software-factory-it-can-trust/https://islo.dev/blog/how-obert-built-a-software-factory-it-can-trust/How Obert uses Islo to turn well-scoped product work into trusted agent workflows and give a small AI-native team more engineering capacity.Thu, 23 Jul 2026 00:00:00 GMThttps://islo.dev/blog/how-we-stopped-working-for-our-agents/https://islo.dev/blog/how-we-stopped-working-for-our-agents/We automated implementation first and it backfired. Starting with review and verification instead, and wiring it all together with webhooks and label-driven triggers, gave us an autonomous engineering loop where a Linear label goes in and a merged PR comes out.Wed, 15 Jul 2026 00:00:00 GMThttps://islo.dev/blog/reward-hack-bench-sandbox-stops-agent-cheating/https://islo.dev/blog/reward-hack-bench-sandbox-stops-agent-cheating/Introducing RewardHackBench, an open-source benchmark for measuring whether sandbox policies prevent reward hacking in agent evaluations.Mon, 15 Jun 2026 00:00:00 GMThttps://islo.dev/blog/agent-sandbox-why-should-i-use-it/https://islo.dev/blog/agent-sandbox-why-should-i-use-it/A technical look at what Agent Sandbox solves inside Kubernetes, what it leaves to the platform team, and how to decide if it's the right primitive for your setup.Tue, 26 May 2026 00:00:00 GMThttps://islo.dev/blog/coding-agents-need-new-primitive/https://islo.dev/blog/coding-agents-need-new-primitive/Fast boot times are useful for request-shaped compute. Coding agents need something closer to a secure, persistent, workstation-shaped runtime.Sun, 17 May 2026 00:00:00 GMThttps://islo.dev/blog/four-layers-of-software-factories/https://islo.dev/blog/four-layers-of-software-factories/A framework for evaluating what you need to run AI coding agents at scale and where most teams get stuck.Mon, 30 Mar 2026 00:00:00 GMThttps://islo.dev/blog/how-we-accidentally-built-doubleagent/https://islo.dev/blog/how-we-accidentally-built-doubleagent/Stateful fake servers for GitHub, Slack, and Stripe that work with official SDKs - built for AI agent development.Tue, 17 Feb 2026 00:00:00 GMThttps://islo.dev/blog/why-islo-loves-monty/https://islo.dev/blog/why-islo-loves-monty/Monty sandboxes runtime-generated Python. Islo gives coding agents a zero-setup environment. Here's what Monty blocks, what can still crash, and how the layers fit.Tue, 10 Feb 2026 00:00:00 GMThttps://islo.dev/blog/competition-docker-daytona-e2b-sprites-coder/https://islo.dev/blog/competition-docker-daytona-e2b-sprites-coder/Here are five sandbox options that work with Claude Code (or with AI coding agents in general), and how they compare.Sun, 08 Feb 2026 00:00:00 GMThttps://islo.dev/blog/understanding-sandbox-security/https://islo.dev/blog/understanding-sandbox-security/Why sandboxing matters for AI agents - isolation, boundaries, and blast radius so you can run agents unattended without risking credentials or production.Sun, 08 Feb 2026 00:00:00 GMThttps://islo.dev/blog/how-secrets-can-be-stolen-clawdbot/https://islo.dev/blog/how-secrets-can-be-stolen-clawdbot/How I sent my own database credentials to a server I control using Clawdbot - and why prompt injection and full machine access make that possible in about 20 minutes.Sat, 07 Feb 2026 00:00:00 GMT