# Context: Plan — Harness
- URL: https://zimmy.dev/harness/plan
- Tags: harness, coding-agent, plan

## User intent hint
Use this focused section as execution context and keep recommendations concrete.

## Hyperlinks (1-depth)
- Compound Engineering: https://every.to/guides/compound-engineering
- SDD Tools — Exploring Gen AI: https://martinfowler.com/articles/exploring-gen-ai/sdd-3-tools.html
- Harness Engineering: https://openai.com/index/harness-engineering/
- Effective Harnesses for Long-Running Agents: https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents
- How Every Codes with Agents: https://every.to/chain-of-thought/compound-engineering-how-every-codes-with-agents
- Building a C Compiler with an Agent: https://www.anthropic.com/engineering/building-c-compiler
- The Loop: https://ghuntley.com/loop/

## Article content
This is a focused section from the Harness workflow.

## 02. Plan
Structured implementation strategy with DAG ordering

1. 구현 순서를 DAG로 나타내라고 한다. 병렬 실행을 위해서.
2. Plan이 Spec과 일치하는지 검토한다. 달라지는 경우가 많다.
3. Architecture, Framework 결정, DB Schema 설계 수준의 high level 기술적 결정은 "읽고 이해한다"
4. Verification Plan도 같이 세우라고한다.
   - 백엔드만 테스트하면 되는지 프론트엔드, E2E도 필요한지
   - 성공 조건을 읽는다.

## References
1. [Compound Engineering](https://every.to/guides/compound-engineering) (every.to)
2. [SDD Tools — Exploring Gen AI](https://martinfowler.com/articles/exploring-gen-ai/sdd-3-tools.html) (martinfowler.com)
3. [Harness Engineering](https://openai.com/index/harness-engineering/) (openai.com)
4. [Effective Harnesses for Long-Running Agents](https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents) (anthropic.com)
5. [How Every Codes with Agents](https://every.to/chain-of-thought/compound-engineering-how-every-codes-with-agents) (every.to)
6. [Building a C Compiler with an Agent](https://www.anthropic.com/engineering/building-c-compiler) (anthropic.com)
7. [The Loop](https://ghuntley.com/loop/) (ghuntley.com)