Portability: Requires a Consensus MCP connection, Node.js with

docx

package, and file reading capability for the syllabus. Works in Claude Code CLI natively. In Claude.ai with Consensus MCP + Code Execution + file upload, the workflow is supported.

.docx

• DOCX generation logic is reusable + complex
• Better separation of concerns: skill = orchestration + intelligence; script = mechanical document assembly
• Token-efficient: skill doesn't re-derive layout each run
• Easier to maintain and version

scripts/generate_reading_list.js

• Only use what Consensus returns. Every paper title, author, journal, year, URL must come from this session's tool calls. Training-knowledge papers labeled

  [Not from Consensus — model knowledge]

  and excluded.
• Confirm before moving on. A search isn't complete until response received and inspected.
• Track three counts. Queries sent / papers received / papers cited. Surface in audit summary.
• Surface gaps, don't fill them. Section with one paper + note about limited results > section padded with fabrications.

Provide the syllabus — pick one:

1. File path (PDF, DOCX, text) — I'll read it
2. Pasted content — paste below
3. Image of a printed syllabus — attach the image

Why I'm asking: Each format needs a different reader (PDF / DOCX parser / vision). Picking upfront prevents wasted attempts.

Course audience — pick one:

1. Undergraduate (intro level)
2. Undergraduate (advanced / upper division)
3. Graduate (Masters / early PhD)
4. Graduate (doctoral / advanced)
5. Professional / continuing education
6. Mixed

Why I'm asking: Audience dictates summary jargon level and discussion-question complexity. Undergrad summaries define every term; grad summaries assume technical fluency. Discussion questions for undergrads test analysis; for grads test critique and extension.

references/audience_calibration.md

Year range for papers — pick one:

1. Last 1 year (most recent only)
2. Last 2 years (default — recent + a year of context)
3. Last 5 years (broader, includes foundational recent work)

Why I'm asking: Reading lists go stale fast. 1-year filters keep things fresh; 5-year filters surface foundational recent work that's already standard. Drives the year_min parameter on every Consensus search.

• PDF: use PDF reader; extract text
• DOCX: use pandoc or DOCX parser; extract text
• Text/pasted: read directly
• Image: use vision; extract text

1. Course title + instructor + term
2. Topic list (lecture titles, week-by-week breakdown, etc.)
3. Learning outcomes (if explicit; if missing, infer 3-5 from description)

[inferred]

Proposed sections: [list with item counts]. Pick one:

1. "Looks good — proceed with these sections"
2. "Merge sections [X] and [Y]"
3. "Split section [X] into two"
4. "Add a section for [topic]"
5. "Remove section [X]"

Why I'm asking: Grouping drives search allocation. Wrong grouping wastes the search budget on bad clusters. This is the last cheap moment to correct course before searches consume Consensus calls.

1. Relevance — paper directly addresses the section topic
2. Reviews / meta-analyses — synthesize the field
3. Citation count — established work
4. Applied-domain connection — tied to the course's domain (e.g., engineering vs theory)

• Plain language (calibrated to audience from Q2)
• 2-3 sentences
• Define jargon if undergraduate audience; assume fluency if graduate

• Bloom higher-order (apply / analyze / evaluate)
• Tied to a specific course learning outcome
• Promotes discussion, not just recall

scripts/discussion_question_validator.py

node ../scripts/generate_reading_list.js \
  --input /tmp/syllabus_data.json \
  --output /path/to/reading_list_<course>_<date>.docx

{
  "courseTitle": "string",
  "courseSubtitle": "string",
  "generatedDate": "string",
  "yearRange": "string",
  "introText": "string",
  "learningOutcomes": ["string", ...],
  "sections": [
    {
      "heading": "string",
      "papers": [
        {
          "title": "string",
          "authors": "string",
          "journal": "string",
          "year": number,
          "url": "string",
          "summary": "string",
          "question": "string"
        }
      ]
    }
  ],
  "auditLog": {
    "totalQueriesSent": number,
    "totalPapersReceived": number,
    "totalPapersCited": number,
    "toolConstraints": "string",
    "searchDetails": [
      {
        "section": "string",
        "query": "string",
        "papersReturned": number,
        "papersSelected": number,
        "status": "string"
      }
    ],
    "failures": []
  }
}

• docx

  package require with multi-location fallback
• Title page, intro with Consensus link, learning outcomes box, numbered papers per section

• ExternalHyperlink

  with full Consensus URLs (never truncated)

• LevelFormat.BULLET

  for lists (not unicode bullets)
• Footer with generation metadata
• Input validation (missing fields → graceful error)

references/bundled_script_pattern.md

• File path
• Audit summary in chat: "Saved {file}. {N} sections × {M} papers / {K} cited. Plan tier: {tier}."
• Validate:

  python scripts/office/validate.py <docx>

• references/applied_domain_weaving.md

  — search-quality canon (7+ sources)

• references/audience_calibration.md

  — undergrad vs grad summary jargon (7+ sources)

• references/bundled_script_pattern.md

  — why bundle vs inline (7+ sources)

• Parallelizing Consensus calls (rate limit)
• Searching topics without applied-domain angle (poor relevance)
• Padding sections with fabricated entries when Consensus returns thin
• Generic discussion questions ("What did the authors find?")
• Jargon-heavy summaries unsuitable for the course's audience level
• Skipping the group-and-confirm step (wastes searches)
• Truncating Consensus URLs in hyperlinks
• Inlining 300 lines of docx-generation JavaScript in the skill body (use bundled script)

megaprompts/10-syllabus-megaprompt.md