# harnessed.md

> What harness engineering is, why it matters, and who's doing it.

> "The primary job is no longer to write code, but to design environments, specify intent, and build feedback loops that allow agents to do reliable work." — [OpenAI, Harness Engineering](https://openai.com/index/harness-engineering/)

## What's harness engineering?

Engineering has fundamentally changed.

Traditionally engineers spent 80% of their time on features, 20% on the system. Harnessed companies flip that on its head: 80% goes to the harness — the machine that builds the machine — 20% guiding where it goes.

The harness has three parts:

- **[Guides](/guides/llms.txt)** — steer the agent _before_ it acts: AGENTS.md, design docs, architecture maps, rules, learnings
- **[Verification](/verification/llms.txt)** — checks the work before it ships: types, linters, tests, agentic review
- **[Observation](/observation/llms.txt)** — monitors what shipped: errors, usage, agentic investigation

```
Intent ◄········································╮
  │            improvements + fixes             ·
  ▼                                             ·
Guides ◄································┬···╮   ·
  │                              evolve ·   ·   ·
  │ constrain + direct                  ·   ·   ·
  ▼                                     ·   ·   ·
Agent builds ◄───────┐                  ·   ·   ·
  │                  │                  ·   ·   ·
  ▼                  │                  ·   ·   ·
Verification ········╁··················╯   ·   ·
  │                  │                      ·   ·
  ▼          no: fix │                      ·   ·
Pass? ───────────────┘                      ·   ·
  │                                         ·   ·
  │ yes                                     ·   ·
  ▼                                         ·   ·
Ship                                        ·   ·
  │                                         ·   ·
  ▼                                         ·   ·
Observation ································┴···╯
```

The system improves itself. Signals from verification and observation loop back into the guides — you don't just fix the code, you refine the harness to stop it breaking that way again.

The [audit](/audit/llms.txt) is a rubric your agent runs against your repo — a one-line prompt, a 0–5 score per item, and a recommended next step grounded in the tooling you already use.

## Sections

- [Guides](/guides/llms.txt) — configuring your agent harness
- [Verification](/verification/llms.txt) — verifying what an agent ships
- [Observation](/observation/llms.txt) — monitoring what shipped
- [Audit](/audit/llms.txt) — score how harnessed your codebase is
- [Jobs](/jobs/llms.txt) — engineering jobs at harnessed companies
- [Companies](/companies/llms.txt) — who is doing harness engineering

## Companies

- [OpenAI](/companies/openai/llms.txt) — 1M lines of code, zero written by humans
- [Anthropic](/companies/anthropic/llms.txt) — Multi-agent harness for long-running coding sessions
- [Shopify](/companies/shopify/llms.txt) — "2026 belongs to those who master agentic harnesses"
- [Every](/companies/every/llms.txt) — 5 products, single-person engineering teams

---

For the complete content in a single file, see [/llms-full.txt](/llms-full.txt)