https://newruntime.com/feeds/agent-runtime.xml The runtime layer around agents: sandboxes, tools, hosted execution, permissions, state, and deployment surfaces. This bounded feed exists because the topic is a featured, evidence-dense New Runtime hub. 2026-09-01T00:00:00.000Z New Runtimehttps://newruntime.com/ New Runtime tracks the systems, signals, and shifts reshaping software as agents become first-class participants in work, products, and organizations. https://newruntime.com/patterns/harness-architecture-outlives-model-choice/ 2026-09-01T00:00:00.000Z 2026-09-01T00:00:00.000Z

For production agents, the harness is becoming a more durable product boundary than the identity of the model running inside it.

https://newruntime.com/projects/github-projects-for-ai-engineers/ 2026-08-15T00:00:00.000Z 2026-08-15T00:00:00.000Z

A Telegram-corpus-derived GitHub radar plus a source-backed map of the models, orchestration, retrieval, memory, security, automation, and infrastructure layers around production AI systems.

https://newruntime.com/posts/google-adk-loop-engineering-state-pruning/ 2026-08-03T00:00:00.000Z 2026-08-03T00:00:00.000Z

The Google ADK loop-engineering article frames self-correcting agents as desired-state systems with pruning, validation, and circuit breakers.

https://newruntime.com/posts/when-machine-consumption-became-the-leading-machine-route/ 2026-07-31T00:00:00.000Z 2026-07-31T00:00:00.000Z

A public methods note on turning an unusual crawler signal into a verified discovery graph, a privacy-safe measurement system, and three falsifiable experiments.

https://newruntime.com/shifts/owned-learning-loop/ 2026-07-29T00:00:00.000Z 2026-07-29T00:00:00.000Z

Organizations are moving the durable learning asset out of one model provider and into user-owned traces, evaluations, corrections, and promotion rules.

https://newruntime.com/posts/hermes-agent-learning-loop/ 2026-07-23T00:00:00.000Z 2026-07-23T00:00:00.000Z

Nous Research's Hermes Agent packages memory, skills, messaging, scheduling, tool use, and sandboxing as a runtime rather than a single chat surface.

https://newruntime.com/posts/devin-outposts-agent-execution-plane/ 2026-07-22T00:00:00.000Z 2026-07-22T00:00:00.000Z

Devin Outposts moves command execution, repository access, and sandbox lifecycle into customer-controlled infrastructure while the agent loop remains in Cognition's cloud.

https://newruntime.com/signals/render-packages-deployment-primitives-for-agent-workloads/ 2026-07-11T00:00:00.000Z 2026-07-11T00:00:00.000Z

Render combines services, workers, databases, cron jobs, previews, and infrastructure definitions into a deployment surface suitable for long-running agent systems.

https://newruntime.com/signals/agentos-lightweight-runtime-for-coding-agents/ 2026-06-29T00:00:00.000Z 2026-06-29T00:00:00.000Z

Rivet's agentOS combines isolated execution, durable workflows, resource limits, and host-side secrets in a WebAssembly and Rust runtime.

https://newruntime.com/signals/deerflow-2-0-packages-an-open-runtime-for-autonomous-agents/ 2026-06-28T00:00:00.000Z 2026-06-28T00:00:00.000Z

ByteDance's open runtime combines planning, subagents, files, tools, and long-running execution into a deployable system.

https://newruntime.com/signals/ai-sdk-7-grows-into-an-agent-runtime/ 2026-06-27T00:00:00.000Z 2026-06-27T00:00:00.000Z

Vercel expands a model integration library with tool loops, state, transport, and UI primitives needed for production agent applications.

https://newruntime.com/signals/the-computer-becomes-a-runtime-assigned-to-the-agent/ 2026-06-19T00:00:00.000Z 2026-06-19T00:00:00.000Z

Giving each agent an isolated machine turns browser, shell, files, and installed software into controlled execution infrastructure.

https://newruntime.com/signals/tokens-stay-the-cost-while-runtime-becomes-the-product/ 2026-06-18T00:00:00.000Z 2026-06-18T00:00:00.000Z

The differentiated layer shifts toward execution, tools, policy, state, and user workflow while model inference remains metered input.

https://newruntime.com/signals/the-agent-becomes-a-separate-hosting-layer/ 2026-06-16T00:00:00.000Z 2026-06-16T00:00:00.000Z

Anthropic's Agent SDK hosting pattern separates model calls from a persistent process that owns tools, state, execution, and recovery.

https://newruntime.com/signals/asteroid-task-specific-ai-interfaces/ 2026-05-21T00:00:00.000Z 2026-05-21T00:00:00.000Z

The archive captures Asteroid as a dated public record from Asteroid. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/google-gemini-models-agent-ready-software/ 2026-05-20T00:00:00.000Z 2026-05-20T00:00:00.000Z

The archive captures Google / Gemini Models as a dated public record from Google / Gemini Models. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as supporting evidence for the agent-ready software trend.

https://newruntime.com/signals/cline-sdk-exposes-an-open-harness-for-coding-agents/ 2026-05-14T00:00:00.000Z 2026-05-14T00:00:00.000Z

Cline separates the reusable agent runtime from its interface, making tool execution and coding workflows embeddable in other products.

https://newruntime.com/signals/claude-managed-agents-add-outcomes-and-multi-agent-orchestration/ 2026-05-07T00:00:00.000Z 2026-05-07T00:00:00.000Z

Anthropic's managed-agent layer adds long-running outcomes, orchestration, and background work around the model call.

https://newruntime.com/signals/github-openai-realtime-voice-component-agent-ready-software/ 2026-04-29T00:00:00.000Z 2026-04-29T00:00:00.000Z

The archive captures GitHub / openai/realtime-voice-component as a dated public record from GitHub / openai/realtime-voice-component. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as supporting evidence for the agent-ready software trend.

https://newruntime.com/signals/google-cloud-gemini-enterprise-agent-platform-ai-native-operating-models/ 2026-04-24T00:00:00.000Z 2026-04-24T00:00:00.000Z

The archive captures Google Cloud / Gemini Enterprise Agent Platform as a dated public record from Google Cloud / Gemini Enterprise Agent Platform. It documents AI adoption shifting jobs, coordination, review, and organizational capacity and is retained as pressure-testing evidence for the AI-native operating models trend.

https://newruntime.com/signals/microsoft-new-hosted-agents-in-foundry-agent-service-verification-bandwidth/ 2026-04-24T00:00:00.000Z 2026-04-24T00:00:00.000Z

The archive captures Microsoft / New Hosted Agents In Foundry Agent Service as a dated public record from Microsoft / New Hosted Agents In Foundry Agent Service. It documents evaluation, review, and observability becoming the bottleneck after generation accelerates and is retained as pressure-testing evidence for the verification bandwidth trend.

https://newruntime.com/signals/openai-routable-model-components/ 2026-04-24T00:00:00.000Z 2026-04-24T00:00:00.000Z

The archive captures OpenAI as a dated public record from OpenAI. It documents models becoming replaceable or specialized components inside a more durable runtime and is retained as supporting evidence for the routable model components trend.

https://newruntime.com/signals/openai-workspace-agents-goal-scoped-agent-loops/ 2026-04-23T00:00:00.000Z 2026-04-23T00:00:00.000Z

The archive captures OpenAI / Workspace Agents as a dated public record from OpenAI / Workspace Agents. It documents long-running work gaining explicit goals, state, stopping rules, and recovery and is retained as supporting evidence for the goal-scoped agent loops trend.

https://newruntime.com/signals/anthropic-claude-design-anthropic-labs-task-specific-ai-interfaces/ 2026-04-21T00:00:00.000Z 2026-04-21T00:00:00.000Z

The archive captures Anthropic / Claude Design Anthropic Labs as a dated public record from Anthropic / Claude Design Anthropic Labs. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/venturebeat-salesforce-launches-headless-to-turn-its-entire-agent-protocol-interoperability/ 2026-04-21T00:00:00.000Z 2026-04-21T00:00:00.000Z

The archive captures Venturebeat / Salesforce Launches Headless To Turn Its Entire as a dated public record from Venturebeat / Salesforce Launches Headless To Turn Its Entire. It documents agent tools and services converging on explicit interoperability protocols and is retained as branch-opening evidence for the agent protocol interoperability trend.

https://newruntime.com/signals/google-ai-mode-chrome-task-specific-ai-interfaces/ 2026-04-20T00:00:00.000Z 2026-04-20T00:00:00.000Z

The archive captures Google / Ai Mode Chrome as a dated public record from Google / Ai Mode Chrome. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/openai-next-evolution-of-agents-sdk-agent-ready-software/ 2026-04-16T00:00:00.000Z 2026-04-16T00:00:00.000Z

The archive captures OpenAI / Next Evolution Of Agents Sdk as a dated public record from OpenAI / Next Evolution Of Agents Sdk. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/github-farzaa-clicky-task-specific-ai-interfaces/ 2026-04-13T00:00:00.000Z 2026-04-13T00:00:00.000Z

The archive captures GitHub / farzaa/clicky as a dated public record from GitHub / farzaa/clicky. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/github-langchain-ai-deepagents-agent-ready-software-2003/ 2026-04-12T00:00:00.000Z 2026-04-12T00:00:00.000Z

The archive captures GitHub / langchain-ai/deepagents as a dated public record from GitHub / langchain-ai/deepagents. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/openai-next-phase-of-enterprise-ai-ai-native-operating-models/ 2026-04-10T00:00:00.000Z 2026-04-10T00:00:00.000Z

The archive captures OpenAI / Next Phase Of Enterprise Ai as a dated public record from OpenAI / Next Phase Of Enterprise Ai. It documents AI adoption shifting jobs, coordination, review, and organizational capacity and is retained as supporting evidence for the AI-native operating models trend.

https://newruntime.com/signals/anthropic-managed-agents-agent-ready-software/ 2026-04-09T00:00:00.000Z 2026-04-09T00:00:00.000Z

The archive captures Anthropic / Managed Agents as a dated public record from Anthropic / Managed Agents. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/github-mattmireles-gemma-tuner-multimodal-generative-media-infrastructure/ 2026-04-09T00:00:00.000Z 2026-04-09T00:00:00.000Z

The archive captures GitHub / mattmireles/gemma-tuner-multimodal as a dated public record from GitHub / mattmireles/gemma-tuner-multimodal. It documents image, video, audio, and multimodal generation becoming application infrastructure and is retained as branch-opening evidence for the generative media infrastructure trend.

https://newruntime.com/signals/github-vercel-labs-agent-browser-task-specific-ai-interfaces/ 2026-04-09T00:00:00.000Z 2026-04-09T00:00:00.000Z

The archive captures GitHub / vercel-labs/agent-browser as a dated public record from GitHub / vercel-labs/agent-browser. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/kaggle-verification-bandwidth/ 2026-04-09T00:00:00.000Z 2026-04-09T00:00:00.000Z

The archive captures Kaggle as a dated public record from Kaggle. It documents evaluation, review, and observability becoming the bottleneck after generation accelerates and is retained as pressure-testing evidence for the verification bandwidth trend.

https://newruntime.com/signals/github-allenai-molmoweb-verification-bandwidth/ 2026-04-05T00:00:00.000Z 2026-04-05T00:00:00.000Z

The archive captures GitHub / allenai/molmoweb as a dated public record from GitHub / allenai/molmoweb. It documents evaluation, review, and observability becoming the bottleneck after generation accelerates and is retained as pressure-testing evidence for the verification bandwidth trend.

https://newruntime.com/signals/olshansky-dear-substack-you-are-missing-out-agent-economics-and-completed-work/ 2026-04-04T00:00:00.000Z 2026-04-04T00:00:00.000Z

The archive captures Olshansky / Dear Substack You Are Missing Out as a dated public record from Olshansky / Dear Substack You Are Missing Out. It documents AI economics moving from token access toward task cost, capacity, and completed outcomes and is retained as supporting evidence for the agent economics and completed work trend.

https://newruntime.com/signals/github-agent-infra-sandbox-agent-ready-software/ 2026-04-01T00:00:00.000Z 2026-04-01T00:00:00.000Z

The archive captures GitHub / agent-infra/sandbox as a dated public record from GitHub / agent-infra/sandbox. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/chrlschn-mcp-is-dead-long-live-mcp-agent-protocol-interoperability/ 2026-03-26T00:00:00.000Z 2026-03-26T00:00:00.000Z

The archive captures Chrlschn / Mcp Is Dead Long Live Mcp as a dated public record from Chrlschn / Mcp Is Dead Long Live Mcp. It documents agent tools and services converging on explicit interoperability protocols and is retained as branch-opening evidence for the agent protocol interoperability trend.

https://newruntime.com/signals/cursor-agent-ready-software-1682/ 2026-03-23T00:00:00.000Z 2026-03-23T00:00:00.000Z

The archive captures Cursor as a dated public record from Cursor. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/github-agent-field-agentfield-agent-ready-software/ 2026-03-20T00:00:00.000Z 2026-03-20T00:00:00.000Z

The archive captures GitHub / Agent-Field/agentfield as a dated public record from GitHub / Agent-Field/agentfield. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/openai-gpt-mini-nano-ai-native-operating-models/ 2026-03-20T00:00:00.000Z 2026-03-20T00:00:00.000Z

The archive captures OpenAI / Gpt Mini Nano as a dated public record from OpenAI / Gpt Mini Nano. It documents AI adoption shifting jobs, coordination, review, and organizational capacity and is retained as pressure-testing evidence for the AI-native operating models trend.

https://newruntime.com/signals/github-thesysdev-openui-agent-ready-software/ 2026-03-16T00:00:00.000Z 2026-03-16T00:00:00.000Z

The archive captures GitHub / thesysdev/openui as a dated public record from GitHub / thesysdev/openui. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as supporting evidence for the agent-ready software trend.

https://newruntime.com/signals/google-ai-gemini-api-agent-protocol-interoperability/ 2026-02-28T00:00:00.000Z 2026-02-28T00:00:00.000Z

The archive captures Google AI / Gemini Api as a dated public record from Google AI / Gemini Api. It documents agent tools and services converging on explicit interoperability protocols and is retained as branch-opening evidence for the agent protocol interoperability trend.

https://newruntime.com/signals/github-vxcontrol-pentag-agent-ready-software/ 2026-02-27T00:00:00.000Z 2026-02-27T00:00:00.000Z

The archive captures GitHub / vxcontrol/pentag as a dated public record from GitHub / vxcontrol/pentag. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/claude-cowork-plugins-across-enterprise-agent-ready-software/ 2026-02-26T00:00:00.000Z 2026-02-26T00:00:00.000Z

The archive captures Claude / Cowork Plugins Across Enterprise as a dated public record from Claude / Cowork Plugins Across Enterprise. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as supporting evidence for the agent-ready software trend.

https://newruntime.com/signals/cursor-agent-computer-use-agent-ready-software/ 2026-02-26T00:00:00.000Z 2026-02-26T00:00:00.000Z

The archive captures Cursor / Agent Computer Use as a dated public record from Cursor / Agent Computer Use. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.

https://newruntime.com/signals/openai-frontier-alliance-partners-ai-native-operating-models/ 2026-02-26T00:00:00.000Z 2026-02-26T00:00:00.000Z

The archive captures OpenAI / Frontier Alliance Partners as a dated public record from OpenAI / Frontier Alliance Partners. It documents AI adoption shifting jobs, coordination, review, and organizational capacity and is retained as supporting evidence for the AI-native operating models trend.

https://newruntime.com/signals/openinterpreter-task-specific-ai-interfaces/ 2026-02-26T00:00:00.000Z 2026-02-26T00:00:00.000Z

The archive captures Openinterpreter as a dated public record from Openinterpreter. It documents model output moving from plain answers into generated task interfaces and actions and is retained as supporting evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/techcrunch-meta-ai-security-researcher-said-openclaw-agent-task-specific-ai-interfaces/ 2026-02-26T00:00:00.000Z 2026-02-26T00:00:00.000Z

The archive captures Techcrunch / Meta Ai Security Researcher Said Openclaw Agent as a dated public record from Techcrunch / Meta Ai Security Researcher Said Openclaw Agent. It documents model output moving from plain answers into generated task interfaces and actions and is retained as pressure-testing evidence for the task-specific AI interfaces trend.

https://newruntime.com/signals/openai-lockdown-mode-elevated-risk-labels-in-chatgpt-agent-security-runtime-boundaries/ 2026-02-19T00:00:00.000Z 2026-02-19T00:00:00.000Z

The archive captures OpenAI / Lockdown Mode Elevated Risk Labels In Chatgpt as a dated public record from OpenAI / Lockdown Mode Elevated Risk Labels In Chatgpt. It documents agent security expanding from prompt policy into memory, tools, sandboxes, and execution boundaries and is retained as branch-opening evidence for the agent security runtime boundaries trend.