|
4 | 4 |
|
5 | 5 | ## Quick API and Automation Commands |
6 | 6 |
|
7 | | -- Run full test suite with uv: `uv run pytest` |
| 7 | +- Install development dependencies: `python -m pip install -r requirements-dev.lock.txt` |
| 8 | +- Run every CI check (lint, tests, exports, examples, courses, paths): `make check` |
| 9 | +- Run the test suite only: `python -m pytest` |
8 | 10 | - Verify runnable examples: `python tools/verify_examples.py` |
9 | 11 | - Verify course folders: `python tools/verify_courses.py` |
10 | 12 | - Validate and inspect resource catalog: `python tools/validate_catalog.py` |
|
19 | 21 | - `catalog/projects/`: One reviewed YAML file per Project Radar entry (discovery candidates are produced on demand by `tools/radar_scan.py`). |
20 | 22 | - `courses/`: Challenge course folders — solve the `TASK.md` contract with a coding agent and prove it with `verify.py` claim codes; `COURSE.md` still offers a guided, agent-taught mode: |
21 | 23 | - `courses/hands-on-python-with-claude-code/`: Task contract, bounded change, and objective verify.py evidence with Claude Code (EN+ZH lessons, three scenario skins). |
22 | | - - `courses/hands-on-with-openai-codex-cli/`: The same verified workflow taught with the OpenAI Codex CLI (EN+ZH lessons, reuses the flagship scenario skins). |
| 24 | + - `courses/hands-on-with-openai-codex/`: The same verified workflow taught hands-on in the Codex desktop app, where an agent thread runs the loop and you supervise diffs and approvals (EN+ZH). |
| 25 | + - `courses/hands-on-with-cursor/`: The same verified workflow taught hands-on in Cursor, where Agent mode runs the loop and you supervise diffs and project rules (EN+ZH). |
| 26 | + - `courses/hands-on-with-deepseek-harness/`: The same verified workflow taught hands-on in DeepSeek Harness, where the agent loop is composed from plugins and every step lands in an inspectable trajectory (EN+ZH). |
| 27 | + - `courses/hands-on-with-kimi-code/`: The same verified workflow taught hands-on in Kimi Code, where built-in subagents split read-only exploration, planning, and the edits you approve (EN+ZH). |
| 28 | + - `courses/hands-on-with-zcode/`: The same verified workflow taught hands-on in the ZCode desktop workspace, where Goal Mode decomposes the work and the Git state shows every change (EN+ZH). |
23 | 29 | - `courses/agent-rules-single-source/`: Build a checker that proves one AGENTS.md source of truth across CLAUDE.md and .cursorrules (EN+ZH, three drift scenarios). |
24 | 30 | - `courses/verifying-ai-generated-code/`: Build the release-evidence machine: run checks, parse results, refuse zero-test projects, write SHIP-RECORD.json (EN+ZH, three project skins). |
25 | 31 | - `courses/mcp-server-in-python/`: Build a stateless 2026-07-28 MCP tool server in pure Python — reuses the reviewed mcp-server example core (EN+ZH). |
|
0 commit comments