---
schema_version: "newruntime-topic-hub-v0.1"
type: "topic_hub"
slug: "harness-engineering"
title: "Harness Engineering - New Runtime"
description: "A New Runtime topic hub collecting signals, patterns, field notes, and public sources about harness engineering."
answer: ["Harness Engineering is tracked here as an evidence-linked topic, not as a static glossary entry.","The page connects raw observations to pattern hypotheses, longer analysis, and public sources.","Use it as the canonical landing page before drilling into individual records."]
search_intents: ["harness engineering","harness engineering AI agents","harness engineering software"]
status: "tracked"
last_updated: "2026-07-16"
counts: {"total":6,"signals":5,"patterns":1,"posts":0,"atlas":0,"sources":11}
routes: {"html":"https://newruntime.com/topics/harness-engineering/","markdown":"https://newruntime.com/topics/harness-engineering.md","json":"https://newruntime.com/topics/harness-engineering.json"}
top_sources: ["https://academy.langchain.com/courses/foundation-introduction-to-deepagents","https://docs.inference.net/introduction","https://github.com/langchain-ai/deepagents","https://github.com/NousResearch/hermes-agent","https://github.com/rivet-dev/agentos","https://modelcontextprotocol.io/docs/concepts/architecture","https://opencode.ai/docs/cli","https://opencode.ai/docs/models","https://opencode.ai/docs/rules","https://opencode.ai/docs/skills","https://x.com/trq212/status/2061907337154367865"]
---

# Harness Engineering - New Runtime

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about harness engineering.

## Short Answer

- Harness Engineering is tracked here as an evidence-linked topic, not as a static glossary entry.
- The page connects raw observations to pattern hypotheses, longer analysis, and public sources.
- Use it as the canonical landing page before drilling into individual records.

## Patterns

- [Harness architecture outlives model choice](https://newruntime.com/patterns/harness-architecture-outlives-model-choice/): For production agents, the harness is becoming a more durable product boundary than the identity of the model running inside it.

## Field Notes


## Recent Raw Signals

- 2026-07-16: [OpenCode makes the model a swappable dependency](https://newruntime.com/signals/opencode-model-swappable-agent-harness/)
- 2026-07-15: [A local agent is a runtime, not one model](https://newruntime.com/signals/local-agent-is-a-runtime-not-one-model/)
- 2026-07-10: [Deep Agents packages a harness for long-running tasks](https://newruntime.com/signals/deep-agents-package-long-running-harness/)
- 2026-07-01: [Deep Agents combines plans, files, memory, and subagents](https://newruntime.com/signals/deep-agents-plan-files-memory-subagents/)
- 2026-06-18: [Claude Code Can Build a Task-Specific Harness Before Coding](https://newruntime.com/signals/claude-code-can-build-a-task-specific-harness-before-coding/)

## Public Sources

- https://academy.langchain.com/courses/foundation-introduction-to-deepagents
- https://docs.inference.net/introduction
- https://github.com/langchain-ai/deepagents
- https://github.com/NousResearch/hermes-agent
- https://github.com/rivet-dev/agentos
- https://modelcontextprotocol.io/docs/concepts/architecture
- https://opencode.ai/docs/cli
- https://opencode.ai/docs/models
- https://opencode.ai/docs/rules
- https://opencode.ai/docs/skills
- https://x.com/trq212/status/2061907337154367865