AI-Shifu Course Creator

Use these capabilities to complete the user's course goal. The using agent owns task decomposition, capability selection, execution order, and teaching design. Start from the available topic, brief, source material, outline, Prompt, or course directory; no questionnaire, segmentation run, intermediate JSON, or separate execution plan is a universal prerequisite.

Reuse existing answers and artifacts. Ask when missing information materially affects the result; otherwise make reasonable design choices within the user's constraints. Capabilities can be combined, revisited, or used alone. A reference explains how to perform its capability, not what the agent must do next. Read the relevant guidance as needed rather than loading a transitive reference graph.

Shared Conventions

• Apply language-policy.md to conversation and authored content, including target-language resolution and terminology.
• Apply session-controls.md to contact eligibility, explicit-request-only version checks, progress, errors, and completion handoffs. First-response contact conditions remain applicable. The completion handoff applies to the user's completed request, not each internal capability call.
• Use Markdown links [descriptive text](URL) for URLs in user-visible messages. URLs inside Teaching Prompts follow MarkdownFlow rules; URLs inside fenced code blocks are exempt.
• Honor explicit author choices, source facts, selected exact material, and the distinction between a per-lesson Teaching Prompt and a course-wide Course Prompt. Runtime semantics are defined in prompt-semantics.md and markdownflow-syntax.md.
• For an existing platform course's lesson, Course Prompt, or description edit, apply the versioned editing baseline before preparing revised content. Course-wide artifact writes use course management with that baseline. Local-only authoring and read-only review remain independent of platform access.

Capabilities

Capabilities fall into three categories: teaching design, expressing that design in MarkdownFlow, and operating the CLI. These categories describe responsibilities, not execution phases. Use one alone, combine them, or revisit them in any useful order. Inputs describe useful starting points, not mandatory upstream outputs.

Teaching Design

Decide what learners should understand or do and how teaching can help them get there. Use existing choices directly; resolve missing choices within the task without requiring a separate design artifact.

Capability	Useful inputs	Result and guidance
Resolve course design	Topic, learner needs, source material, existing choices	Design choices and their effects; ask only material unresolved questions. Course design
Analyze material and plan structure	Source material, topic, outline, existing lessons	Relevant source analysis, chapter and lesson structure, or optional traceable segments. Structure and segmentation
Design lesson teaching	Goals, audience, source, existing teaching choices	Suitable explanations, examples, practice, interactions, visual purposes, and feedback. Teaching options
Write listing copy	Course topic, audience, goals, and supported outcomes	Learner-facing course description; this prose does not require MarkdownFlow. Course description
Review teaching quality	Available material, design, or course content	Scoped findings and requested repairs for factual support, source coverage, and learning effectiveness. Review, checks

Teaching patterns, interaction counts, and visual rhythms in teaching-design.md are recommendations. The agent chooses teaching strategy within the user's explicit constraints.

Express in MarkdownFlow

Make the intended teaching behavior executable in Teaching Prompts and Course Prompts. Preserve supplied design choices; use the teaching-design capabilities for choices that remain open rather than letting syntax or a layout template determine pedagogy.

Capability	Useful inputs	Result and guidance
Write or revise a lesson	Topic, source, existing Prompt, or supplied design	Runnable Teaching Prompt with required source-only explanatory comments and a suitable layout. A separate design run is not required. Teaching Prompt
Write or revise a Course Prompt	Course title, topic, goals, audience, and design context	Course-wide Prompt using the existing six-section template. Course Prompt
Encode interactions, variables, and exact content	Intended response behavior, variable uses, selected exact material	Valid runtime controls, references, branches, and deterministic output. MarkdownFlow, Prompt semantics
Express lesson images	Teaching intent, authoritative asset URLs and metadata	Correct image embedding or HTML-view instructions. Image authoring
Review Prompt expression	One Prompt or selected artifacts	Scoped findings and repairs for runtime syntax, variables, comments, layout, and artifact contracts. Review, checks

Operate the CLI

Use deterministic commands for local course files, platform access and changes, and observed course data. Existing Prompts and directories can be used directly without redoing teaching design or Prompt authoring. The CLI reference owns command mechanics; the course-directory specification owns recognized files, build precedence, and import format.

Capability	Useful inputs	Result and guidance
Establish platform access	Selected service/profile and existing authorization	Verified access and recoverable browser authorization when needed. Authentication
Identify an existing course	BID or title and selected platform context	Unique target or an explicit unresolved result. Course target
Upload lesson images	Image assets, informative alt text, and selected platform context	Real resource URLs and provenance records for later embedding. Image upload, asset handling
Synchronize existing content	Existing course and local edits/directory	Pull, divergence inspection, version-protected writes, and conflict convergence. Course sync
Build or import a new course	Authored course directory or import payload	Built payload, new draft, and verified imported content. Build and import
Manage and publish a course	Resolved course and requested operation	Metadata, structure, access, avatar, Listen Mode, publication, preview, or archive changes. Course management
Analyze course data	Course/question, dimensions, and time range	Queries and readable, privacy-respecting findings. Analytics

Shared references support the relevant category: authoring-contracts.md defines optional author controls and artifact consistency; source-preservation.md defines exact-source decisions. An image's teaching purpose, MarkdownFlow representation, and CLI upload are separate actions even when they share a reference. Review likewise applies the criteria for the requested design, expression, or platform result.

Local teaching design, Prompt expression, content-only review, and builds do not require platform access; invoke authentication when a selected platform operation needs it. Analytics may inform a later teaching decision when relevant, but querying data does not itself request a course rewrite.

Full-Course Creation Outcome

A request to create a complete AI-Shifu course includes deployment and publication by default. An explicit local-only, unpublished-draft, or narrower deliverable request determines that narrower outcome. A title-only creation request is still course authoring; it does not mean creating an empty platform draft. This outcome does not prescribe the order of authoring capabilities.

Before creating a new platform course, show the completed title, structure, content preview, and selected platform attributes, and obtain confirmation to create and publish that concrete result. Use draft-only wording when the user requested an unpublished draft. Reuse approval of the same completed result and scope; revise and obtain approval again if the approved result changes. Keep local work when confirmation is deferred. Browser login approval is separate from this content confirmation. This checkpoint applies to new-course creation and does not add a confirmation step to existing-course synchronization or independent management requests.

Keep new-course intent distinct from an existing-course edit. A no-match lookup does not authorize creating a replacement course. Reuse the BID of an already-created course when resuming incomplete work. Report the operations and verification that actually completed; an imported draft or a printed public URL alone does not establish successful publication.