Course Creator — beta

course-creator helps an AI agent create a course for a defined learner. It maps what the learner already knows, orders concepts by prerequisite, writes explanations in progressive modules, and links factual statements to reopenable sources. Independent fresh learner agents can expose confusing steps; their results are diagnostic. The course is labeled efficacy not verified until enough evidence from real learners supports a narrower claim.

The package is the fourth lens of the Agentic SDLC family. It uses the shared ai_docs/ project memory and governance. Version 0.1.0-beta.4 supports Standalone mode. This is an experimental beta: the authoring method and artifact contracts may change; human learning efficacy has not been established. Its course artifacts remain readable without the skill. Each new course keeps a verbatim, append-only record of the requester's own words and decisions in audit/requests/, cited by the analysis and checked by its reviews.

What's new in 0.1.0-beta.19

Codex hook checks can recognize the work the agent is doing. The shared validator recognizes supported native Codex patches and transcript records, and gives more accurate orientation-hook diagnostics when definitions are disabled or cannot be verified. Existing hook setup and opt-outs are preserved.

What to do

Run npm install -g @antoneeo/course-creator@latest, then course-creator-install-skill, and restart your agent sessions so they load the updated validator. Check that npm ls -g @antoneeo/course-creator shows 0.1.0-beta.19. Any configured Codex hooks still require their normal client trust review.

Full history: CHANGELOG

Install and initialize

npm install -g @antoneeo/course-creator
course-creator-install-skill
cd YOUR_PROJECT
course-creator-init

The installer detects Claude Code, Gemini CLI, Codex and Google Antigravity. Initialization creates absent ai_docs/ files and a project protocol pointer; it does not overwrite existing project documents. Use python <skill_dir>/scripts/sdlc_check.py check --root . to validate the shared SDLC and course structures.

Docs site (scripts/docs_site.py --root <project> [--open] [--zip <file>]): a static, offline, read-only website of what the agent reads in the project — reading guide, Vision, rules, documents, open work with each next step, courses and, with a knowledge base, the topic map, each fact with its exact source passage, and the source documents. Written to .docs-site/ (added to .gitignore on the first run), outside ai_docs/; nothing is ranked or recommended.

Using the skill in your own project carries no obligation to publish your course or its sources. See LICENSE for the software terms and NOTICE for attribution.

Primary course output

New courses deliver a complete textual source for every slide (SLIDE_CONTENT.md), with exact learner copy, full explanations, examples, visual contents, transitions and answers. PPTX and other renderings are optional. Explicit non-slide requests retain a complete textual source for their units. Every course must show distinguishable value for its audience through an explanatory contribution and a transfer task; structure or simulated success does not establish human efficacy. See skills/course-creator/slide_content.md.

A course delivered by an instructor (Delivery mode: instructor-led) also ships INSTRUCTOR_GUIDE.md: what to prepare, the schedule, what to keep out of the learners' view, what to do when a live step fails, what to clean up afterwards, and the list of every material the delivery needs, wherever it is in the project, checked to exist. course_check refuses an instructor-led course without the guide, or with a guide that does not follow the template in templates.md. An existing instructor-led course therefore fails check until its guide is written to that template. In any course with a content contract it also refuses a line that names the course with a dotted version other than the declared Course version.

Course design now uses outcome-first evidence, teaching/assessment alignment, visible worked reasoning, proportionate support, cognitive-load management and retrieval opportunities. learning_design.md records the pedagogical sources and the source-informed semantic review rubric; a locator table or structural PASS is not proof of learning.