AutonomousVehicleControlBeginnersGuide
Python sample codes and documents about Autonomous vehicle control algorithm. This project can be used as a technical guide book to study the algorithms and the software architectures for beginners.
翻訳待ち
No updates detected yet. GitHub releases will appear here.
Agent-first adaptive learning platform. Create courses as YAML knowledge graphs via CLI or MCP server. AI agents scaffold, fill, review, and publish — no UI required. Adaptive diagnostics, spaced repetition, white-label sites.
翻訳待ち
Loading the full README. The saved excerpt is shown below.
Build courses with lessons, practice questions, and scheduled review.
Graspful stores academies, courses, and brand configuration in YAML files. You or an external agent, such as Claude or Codex, author the content through CLI commands or MCP tools. The CLI creates draft scaffolds. The author supplies the source material, writes the lessons and questions, and reviews the facts before publication.
Learners work through explanations, examples, and questions. Their answers update progress estimates. Prerequisites guide lesson order, and the system schedules later review.Read the full README ↗Build courses with lessons, practice questions, and scheduled review.
Graspful stores academies, courses, and brand configuration in YAML files. You or an external agent, such as Claude or Codex, author the content through CLI commands or MCP tools. The CLI creates draft scaffolds. The author supplies the source material, writes the lessons and questions, and reviews the facts before publication.
Learners work through explanations, examples, and questions. Their answers update progress estimates. Prerequisites guide lesson order, and the system schedules later review.
graspful validate and graspful review. Automated checks report structural problems. The author still checks accuracy and teaching quality.--publish saves a draft.Read the course authoring runbook before building a course.
bun add -g @graspful/cli
# Create a local draft scaffold.
bunx @graspful/cli create academy --topic "Your topic" -o academy.yaml
mkdir -p courses
bunx @graspful/cli create course --topic "Foundations" -o courses/foundations.yaml
These files need authoring and review before import. Follow the runbook to complete the content and match the academy manifest to its course files.
Register before using import or publish commands:
bunx @graspful/cli register --email [email protected]
Registration opens browser authentication, creates a private organization workspace, and saves an API key locally. MCP reads those saved credentials for authenticated tools. GRASPFUL_API_KEY overrides the saved key. A course or brand import creates the public website. See auth and access.
| Command | Auth required | Purpose |
|---|---|---|
graspful register | No | Register through browser authentication and save an API key |
graspful login | No | Authenticate an existing account |
graspful create academy | No | Create an academy plan and manifest scaffold |
graspful create course | No | Create a course YAML scaffold |
graspful create brand | No | Create brand YAML with theme presets |
graspful fill concept | No | Add placeholder knowledge points and questions |
graspful validate | No | Validate the YAML schema and prerequisite graph |
graspful review | No | Run mechanical quality checks |
graspful describe | No | Show course statistics |
graspful import | Yes | Import YAML, saving a draft unless publication is requested and succeeds |
graspful publish | Yes | Request publication of a reviewed draft |
Check the command result to confirm whether publication succeeded.
graspful init authenticates through the browser and configures detected clients. For local authoring before registration, configure the MCP server directly. This example uses the Claude Desktop and Cursor format; see the linked MCP guide for other clients:
{
"mcpServers": {
"graspful": {
"command": "bunx",
"args": ["@graspful/mcp"],
"env": { "GRASPFUL_API_KEY": "gsk_..." }
}
}
}
The external agent writes the course content and uses Graspful tools to scaffold, validate, review, import, and publish it. The MCP server performs structured course operations. See the MCP documentation for setup and the tool reference.
Stripe and Stripe Connect integration code exists. Paid subscriptions and creator payouts are not ready for customers. Live credentials, price configuration, creator onboarding, webhook setup, and payment verification are required before accepting payments.
The repository contains proposed pricing and revenue-share settings. Treat those as configuration under development, rather than an offer to customers.
@graspful/cli@graspful/mcpgraspful/
├── apps/web/ # Public site, academy pages, and learner app (port 3001)
├── apps/site/ # Separate creator dashboard app (port 3002)
├── backend/ # NestJS API
├── packages/
│ ├── creator-ui/ # Components and API contracts shared by both apps
│ ├── client/ # CLI and MCP HTTP client
│ ├── shared/ # Schemas, types, quality checks
│ ├── cli/ # CLI
│ └── mcp/ # MCP server
├── content/
│ ├── courses/ # Course YAML files
│ ├── brands/ # Brand YAML files
│ └── academies/ # Academy manifests
└── docs/ # Documentation
bun install
bun run dev
bun run build
bun run lint
bun run typecheck
bun run test
bun run test:scripts
Copy the relevant environment examples into each workspace before starting services. apps/site provides a dedicated creator deployment and login surface; apps/web serves the platform and branded learner sites. Shared creator code lives in packages/creator-ui.
Run browser tests in both apps with bun run test:e2e. Follow local E2E setup to use an isolated database and auth service. CLAUDE.md lists the test requirements.
Maintenance scripts live in backend/scripts. Run the demo seed from backend with bun run seed:demo electrical or bun run seed:demo javascript. The reference book is available as The Math Academy Way PDF; local downloads are ignored.
See LICENSE.md for the license terms.