Week 4

Skills

Different Models

Connectors (MCP)

Subagents

Practice

Assignment

Back to Track

Content

  1. Model A/B. Take one small task from your project. Run it on a fast model, roll back, run on a powerful one. Post: the task, both results, and where the difference did (or didn't) show. Half the value is learning when the powerful model isn't needed.
  2. Extract a skill. Find an instruction you've typed into chat more than once this month. Turn it into a skill (folder + SKILL.md), then trigger it with a natural request — without naming the skill. Post the SKILL.md and proof it fired.
  3. Build the reviewer. Create the code-reviewer subagent from the topic page (adjust the checks to your project). Run it on your Week 3 PR. Post one issue it found — and one thing it flagged that you disagree with, with your reasoning. The reviewer is input, not verdict: Week 1's rule applies to its output too.
  4. One new connector. Connect a service your capstone will need — via Desktop settings and verify it in the CLI with /mcp. Post what you connected and the first useful thing the agent did with it.

The HackYourFuture curriculum is licensed under CC BY-NC-SA 4.0 *https://hackyourfuture.net/*

CC BY-NC-SA 4.0 Icons

Built with ❤️ by the HackYourFuture community · Thank you, contributors

Found a mistake or have a suggestion? Let us know in the feedback form.