Harness
We track 19 posts about Harness from 13 engineering blogs. Most active: Vercel, SitePoint, Scrapinghub. Latest post: Nov 3, 2026.
Companies writing about Harness
Recent posts
The Agents on Rails Files: enter lemans, a Ruby harness for fair agent races (opens on the source site)
Authors: Vladimir Dementyev, Principal Backend Engineer, Artur Petrov, Backend Engineer, and Travis Turner, Tech Editor Topics: Rails, AI, Open Source, LLM benchmarking, Ruby, LLMs A tour of lemans, the open source Ruby harness behind Agents on Rails: eval task anatomy, how it runs and grades coding agents for hours, and how cheaters shaped its guardrails. Benchmarking coding agents in one sentence: give a model a task, let it work, check the result. The hard part is keeping the race fair. Every model must get the same conditions, and no agent should be able to escape its sandbox or fool its…
Prospero: a new coding harness for a more civilized age (opens on the source site)
I’ve been fascinated with local LLMs for a couple of years now. The better open weight models, including some that inference on regular, higher-end consumer hardware at tolerable speeds, are now reliable enough to handle simpler programming tasks. This requires a coding harness, but so far this year I’ve been disappointed by all the harness offerings. For inexplicable reasons most are built with web stacks, and consume enormous amounts of memory. At its core a coding harness is essentially curl glued to popen(). This should not require gigabytes of RAM. So I built my own: Prospero. It’s…
The fx adapter now runs the harness on the host (opens on the source site)
Vercel ·
The fx adapter for the AI SDK’s HarnessAgent now runs the harness and model requests in the same process as your application. Built-in filesystem and shell operations remain isolated in the sandbox. The adapter uses libfx, the TypeScript library for fx. It no longer needs to install fx inside the sandbox, run a sandbox bridge, or expose a sandbox port. Model credentials stay on the host rather than being forwarded into the sandbox. The rewrite also adds: Built-in tool filtering: choose which built-in tools the agent can use Mid-turn steering: send new instructions while a turn is running…
HarnessAgent now supports image inputs for all harness adapters (opens on the source site)
Vercel ·
The AI SDK HarnessAgent layer now accepts image inputs across every harness adapter. Pass images alongside text in the standard AI SDK message format: a content array mixing { type: 'text' } and { type: 'file' } parts. Image input is processed by the harnesses' underlying vision models and supported across all available adapters: Claude Code, Cline, Codex, Cursor, Deep Agents, fx, GitHub Copilot, Grok Build, OpenCode, Pi. Read the HarnessAgent image inputs documentation to get started. Read more
Making every harness durable (opens on the source site)
Vercel ·
Developers are familiar with running coding agents locally through harnesses like Pi, Claude Code, and others. They are reliable daily drivers for generating code, but keeping agents running around the clock requires your computer to be on and awake, which led to the laptop lid memes and Mac mini demand spike. Running harnesses in the cloud Harness workloads are moving to the cloud. Individual developers are deploying always-on coding agent workflows, projects like OpenWork and HanaAgent use open-source harnesses as the engines behind the apps, and engineers are building their own harness…
AI Agent Harness (opens on the source site)
Birgitta Böckeler uses a similar way of explaining this relationship in her discussions on harness engineering. She describes the LLM as the core and the harness as the surrounding layer that brings together tools, prompts, and other mechanisms that make the model more useful for real work. A better model does not automatically mean a better agent. The environment around it matters too. Why does an AI agent need a harness? A model can reason about a problem and still struggle to complete the task. Take this request: “Fix the authentication bug in this application.” Identifying a possible fix…
Building an evidence-grounded agentic security operations harness on Cloudflare (opens on the source site)
Cloudflare Managed Defense uses a team of specialized AI agents built on Workers and global network telemetry to analyze security alerts. By separating deterministic evidence collection from model inference, the system delivers grounded recommendations to Managed Defense Analysts.
Free the models: Harness design at the frontier (opens on the source site)
Replit ·
Model routers are everywhere right now: a small model reads each turn and picks which model should handle it. The inconvenient truth is that the router is always less capable than the model it’s choosing for. Replit Agent lets the model decide instead. The main agent, or core loop, chooses its subagents’ tier and effort, and adjusts its own as the task unfolds. Given that freedom, Astra hands routine implementation to less costly subagents and decides for itself where its tokens are worth spending. On both DeepSWE and Terminal-Bench, Replit Agent is Pareto-efficient against Astra on its own:…
Securing AI Pull Requests: Building a Deterministic AST Audit Harness in GitHub Actions (opens on the source site)
Build a deterministic CI/CD audit harness that checks AI coding-agent pull requests for AST prototype pollution, secret leakage, and unsafe egress before merge. Continue reading Securing AI Pull Requests: Building a Deterministic AST Audit Harness in GitHub Actions on SitePoint.
Build a Safer AI Agent Harness with Jev and LangChain (opens on the source site)
Build a typed decision layer around a LangChain agent to route simple work to cheaper models and require an explicit risk decision before sensitive tool calls execute. Continue reading Build a Safer AI Agent Harness with Jev and LangChain on SitePoint.
Build Your Own AI Agent Harness in C#, the MafClaw Live Series (opens on the source site)
.NET ·
I am building a complete C# agent live, from a single call around an IChatClient to a production-ready, observable, governed agent, using the Microsoft Agent Framework harness in a 4-part Microsoft Reactor series. The post Build Your Own AI Agent Harness in C#, the MafClaw Live Series appeared first on .NET Blog.
AI SDK harness layer now supports native subscription authentication (opens on the source site)
Vercel ·
The AI SDK harness layer now supports authenticating harnesses through their native subscriptions, where the underlying harness supports them. The harness layer runs different coding agents through the same HarnessAgent interface, so you can switch agents without changing your application code. No code changes or new settings are required. The direct authentication mode uses explicit provider environment credentials when they are present, and otherwise a native subscription found on the host. The default auto mode does the same when no AI Gateway credentials are set. The ai-gateway mode never…
From AI Code to Trusted Software: Harness Engineering in Practice (opens on the source site)
Your team is shipping more AI-written code every quarter. How do you know it is any good?Good means it meets your standards, and you have probably already tried handing your agents the standards: a context file, a style guide, the wiki pasted into the prompt. The agent reads them, agrees, and still breaks them, because instructions to a model are suggestions. What the repository permits is what actually happens. Your people absorb standards through review comments and hallway corrections, and the lessons stick. An agent apologizes and forgets by the next session. The only place its lessons…
Below the Harness: Governing a Multi-Model, Multi-Harness World (opens on the source site)
Docker ·
We believe the future is a multi-model, multi-harness world. And we think it needs a new trust model. In 1988, Norm Hardy described a problem that had been quietly breaking systems for years: the confused deputy. A program that takes action using its permissions instead of yours. Today, every AI agent is that deputy. It...
Lean, GenAI et Delivery Tech - Entretien Sylvain Fagnent : Le Lean est au coeur du sujet du Harness Engineering (opens on the source site)
Sylvain Fagnent est un des experts les plus avancés sur les sujets du Low Code / No Code et de la Gen AI chez OCTO Technology. Il a rédigé de nombreux articles sur notre blog et il nous semblait naturellement un candidat idéal pour réfléchir aux intersections entre la pensée Lean et l’avènement de la GenAI dans les processus de Delivery Tech.
Codex CLI: OpenAI's Open Agent Harness — Installation, Commands, and CI Integration (opens on the source site)
OpenAI released Codex CLI as Apache 2.0 open-source agent harness with interactive CLI, programmatic SDK, and Codex app-server for product embedding. v0.147.0 shipped August 7, 2026. Most articles cover it as chat interface — developers need practical coverage of codex exec for CI pipelines and one-shot tasks. Continue reading Codex CLI: OpenAI's Open Agent Harness — Installation, Commands, and CI Integration on SitePoint.
Harness Engineering, part 4: giving your agent a custom fetch tool that survives the real web (opens on the source site)
Your AI agent is as powerful as the tools it has access to. Here's a tutorial on how you can create your own custom agent tool using Claude Agent SDK for zyte which makes getting structured data from web a breeze.
Building an agentic harness that outlasts the model (opens on the source site)
Shopify ·
We built an agentic code review and test oracle harness that discovers vulnerabilities, proves them with real tests, and provides Shopify-tuned fixes.
Harness Engineering #3- Headless mode: the minimal agent harness (opens on the source site)
What's the smallest harness that still works? Turns out it's already sitting inside almost every coding agent you have installed — headless mode: same loop, tools, and reasoning as the interactive agent, minus the human in the chair. We point it at a web page and pull clean, structured data out the other end in about ten lines.
Related topics
This page is generated automatically from the engineering blogs we follow. Every post links to its source, where it was published. See all sources.