Skip to content

Black Friday: up to 61% off our packs until November 30.

See the packs

Blog

Blog & Insights

Practical articles on building and running AI products, agents and workflows.

All articles

    • Security & Privacy

    OpenCode Browse Safe Setup: Isolating Coding Agents From Your Signed-In Browser

    OpenCode Browse lets coding agents drive your real, signed-in browser. That makes it convenient, and it also makes it risky. This guide sets out an OpenCode Browse safe setup built on the Five-Layer Containment Model: Profile, Scope, Mode, Handoff and Cleanup. It explains what allowedOrigins and readOnly do not protect against and ends with a pre-flight checklist.

    • Security & Privacy

    Protegrity-PII: How to Screen Agent Prompts and Tool Calls for PII, and Re-Test Vendor Benchmarks First

    Protegrity-PII is an open-weights, 278M-parameter PII detector trained on 69 entity types. It scores well on a benchmark Protegrity built itself. This guide covers how it compares with Presidio, Amazon Comprehend and OpenAI Privacy Filter, where to place it around an agent's prompts and tool calls, and how to re-test the vendor's numbers on your own data.

    • Security & Privacy

    IBM Granite 4.1 LLMON: How Instruction/Data Separation Cuts Prompt Injection and How to Test It

    IBM's Granite 4.1 LLMON previews (3B and 8B) use markup to keep trusted instructions apart from untrusted documents and tool outputs. This guide explains how the format works, offers a reusable lens for reading IBM's attack-success numbers, and shows how to test the separation on your own retrieved data before you rely on it.

    • Security & Privacy

    wfx GitHub Actions Secret Stealer: How to Hunt for It When Rotating Keys Isn't Enough

    The wfx GitHub Actions secret stealer used stolen but valid tokens for real developer and bot accounts, opening pull requests and pushing commits the way a developer would. Here is why rotating keys won't close the compromise, and the order to hunt in across GitHub, Entra, AWS and AI providers.

    • Security & Privacy

    Linux Foundation Agent Name Service: How DNS-Anchored AI Agent Identity Works and When to Adopt It

    The Linux Foundation Agent Name Service (ANS) gives AI agents versioned, DNS-anchored names, certificates issued after domain validation, and public transparency logs. This guide covers how it works, where it sits next to DNS-AID and AGNTCY Directory, and how IT leaders can adopt it in stages with the Optijara Anchor-Version-Tier ladder.

    • Security & Privacy

    Fireworks AI Breach: An Inference-Provider Risk Checklist After the LLMjacking Incident

    The Fireworks AI breach involved two intrusions in October 2026, and the known impact grew from evaluation-job secrets to billing data, datasets and incomplete models. This guide uses Fireworks' own timeline to build the Secrets, Data, Spend playbook and a vendor due-diligence checklist for teams running workloads on hosted inference.

    • Security & Privacy

    Claude Code settings.local.json Secret Leaks: Audit 'Don't Ask Again' Rules and Rotate Exposed Keys

    When you choose 'Yes, and don't ask again' in Claude Code, it saves the approved command exactly as typed into settings.local.json inside the project. If the command held a credential, the file now holds it too. Security researchers have found thousands of these secrets in public repositories, Docker images and npm packages. This guide gives engineering leaders a sequence called Find, Kill, Clean, Close: locate every copy, rotate the credentials before cleaning anything, and then shut the paths the file leaks through.

    • Security & Privacy

    Cloudflare Personal Agent Policies: How to Allow, Limit or Block AI Assistants on Your Site

    Cloudflare's personal agent policies let you set a separate access action for each supported AI assistant, see assistant traffic as its own analytics category, and stop one abusive account without blocking an assistant for everyone. This guide explains each control, suggests starting positions for ad-funded, SaaS and ecommerce sites, lays out a five-step rollout loop, and is clear about what the controls cannot see.

Next step

Have something to build?

Tell us what you need. We reply with a clear scope and a fixed quote.