# Task template for an agent

AI Workshop · episode 5: Prompts that work
https://mihalkevich.com/en/workshop/prompts

- [ ] Goal: what should change for the user — one sentence
- [ ] Context: which files, pages, APIs are involved; a link to a similar place in the code
- [ ] Constraints: what not to touch, what style, which libraries not to add
- [ ] Definition of done: the commands and checks that must pass
- [ ] Example: what a correct result looks like (a code fragment, text, a screenshot)
- [ ] Output format: a plan before code; at the end, the list of changed files and check output

## Harness in this episode

A definition of done in the prompt is the bridge to the harness: what you check by eye today becomes an automated test tomorrow.

## Official documentation

- [Prompt engineering — overview](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/overview)
- [Claude prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices)
- [Claude Code — best practices](https://code.claude.com/docs/en/best-practices)

— mihalkevich school