Practice mode · free · unlimited · not certifiable

Local Files and Project Context

Unscored tasks with hints and instant feedback. Practice results never become credentials.

Short lesson

Working with agents safely · 10 min

Name each agent by role and task, isolate them in worktrees, hand off through files, review every diff, never paste secrets. Tool notes (2026-10): Claude Code (CLAUDE.md, agents folder), Codex CLI and others (AGENTS.md), Gemini CLI (GEMINI.md), Cursor, Copilot — verify names in each tool's docs.

L1Point the agent at the right fileterminal

Start a mock agent named docs-summary (profile scout) in ~/shop and ask it to summarize ONLY docs/api.md (use --files).

L3Project instruction file and exclusionsterminal

Prepare ~/shop for AI agents: write a project instruction file (AGENTS.md, with CLAUDE.md as a copy or link) stating what the project is, how to build and test (npm ci, npm test), conventions, and what not to touch (.env, dist/). Add an .agentignore that excludes .env, node_modules/ and dist/. Then start agent builder-validation (profile steady, cwd ~/shop) and run it with the task “add input validation” without --files.