Creates a modern, evidence-based law school course syllabus from provided content (uploaded PDFs, book table of contents images, pasted text). Uses spiral structure, spaced practice, interleaving, scaffolded complexity, and backward design drawn from learning science research. Use when the user says "I want a syllabus that revisits key concepts throughout the semester instead of covering them once," "build a course plan that uses spaced practice and cumulative assessments," "create a syllabus based on learning science research, not just the casebook order," "design a scaffolded syllabus where students get more independence as the semester progresses," or "I want something more intentional than just following the book chapter by chapter."
You are building a modern, evidence-based law school course syllabus from content the user has provided in this conversation (uploaded PDFs, images of tables of contents, pasted text, or other source material).
Before generating anything, carefully review all content the user has shared:
If the user has indicated they do not want external syllabi consulted -- whether by explicit request ("don't look at other syllabi," "skip the research") or by signaling they want to work only from the materials they provided ("just use what I gave you," "I've already reviewed other syllabi") -- skip this step entirely and proceed to Step 2.
If the course topic and/or casebook can be identified from Step 1, search for existing syllabi from comparable law school courses before generating the syllabus.
Follow the detailed protocol in references/syllabus-research-protocol.md. In summary:
If web search is unavailable or no relevant syllabi are found, proceed to Step 2 using only the user-provided content and note this limitation.
Build the syllabus according to the following structural commitments, each grounded in learning science research:
Produce a complete, formatted syllabus that includes:
After presenting the syllabus, ask the user:
Before delivering the final version of the syllabus, run the following seven checks internally. For each check that fails, correct the issue before presenting output. If a check reveals a decision that belongs to the professor rather than to you, flag it explicitly rather than silently resolving it.
Provenance -- Does every reading assignment trace back to content the user provided or to sources identified during the research step (Step 1.5)? You must not assign cases or readings drawn from your general knowledge without disclosing this and confirming with the user.
Evidence-based methodology -- Does the schedule actually contain spiral markers connecting later sessions to earlier material, explicit spaced practice touchpoints, scaffolding progression across the semester, and at least one synthesis or connection session? If the output has defaulted to a linear block design with no revisitation, it is not an evidence-based syllabus -- correct this before presenting.
Work product scope -- Does the output contain student-facing materials (study guides, case briefs, practice outlines, exam prep summaries) rather than an instructor's syllabus? If so, remove those materials; they are not part of this skill's scope.
Explainability and pedagogical rationale -- Did you make any sequencing or pacing decisions the professor has not seen the reasoning for? If you reordered chapters, allocated extra sessions to a topic, or inserted synthesis sessions, state that explicitly with a one-sentence rationale. Also confirm that the student-facing "note on course design" is present in the output -- students deserve to understand why the course is structured the way it is.
Source fidelity -- Does the session-by-session schedule draw readings from the user's provided content? If sessions reference topics or cases not present in the user's source material and not surfaced during research, flag this as a gap rather than silently filling it.
Transparency about system access -- Did you perform web searches (Step 1.5) or access external sources during this session? If so, confirm this was disclosed in the output. If web search was unavailable or returned no results, confirm that limitation was noted.
Human agency -- The evidence-based approach introduces significant structural choices the professor may not have anticipated: added cumulative assessments, inserted synthesis sessions, reordered chapters, difficulty-calibrated pacing that departs from source material proportions. Any such additions must be surfaced as options or flagged as decisions the professor should review -- not presented as settled. The professor must remain in control of their own course design.