Stage-gated Codex skill for doctoral proposal and thesis drafting.

This repository packages a reusable academic workflow for turning rough thesis ideas into aligned research deliverables while preserving truthfulness, traceability, and backbone consistency across title, gap, framework, variables, method, evidence, and chapter outputs.

• Proposal-ready workflow for Chapters 1-3
• Thesis-stage classification from topic selection to viva preparation
• Research-backbone alignment before prose expansion
• Traceable literature-base construction for current or niche topics
• Hard and soft gate checks before calling a draft usable
• Practical work-product orientation instead of advice-only prompting

flowchart LR
    A["User Thesis Request"] --> B["Stage Classifier"]
    B --> C["Backbone Alignment"]
    C --> D["Minimal Literature Base"]
    D --> E["Stage-Specific Deliverable Drafting"]
    E --> F["Hard and Soft Gate Checks"]
    F --> G["Proposal Files / Review Outputs / Revision Matrix"]
    E --> H["Optional Figures-Diagram Skill"]
    H --> G

Use this skill for:

• PhD proposal drafting
• DBA and EdD thesis support
• literature review structuring
• methodology planning
• revision after supervisor feedback
• research-gap clarification

Depending on the request, the skill can help produce:

• refined title options
• research gap statements
• proposal chapters 1-3
• hypotheses or propositions
• methodology plans
• revision matrices
• literature retrieval files

Common companion files include:

• project_config.yaml
• research_plan.md
• references/query_plan.md
• references/candidate_papers.jsonl
• references/evidence_table.csv
• outputs/proposal_draft.md
• outputs/revision_matrix.md

SKILL.md
agents/
  openai.yaml
references/
  gates.md
  literature-retrieval.md
  proposal-workflow.md
scripts/
  semantic_scholar_search.py
assets/
  banner.svg
  preview.svg

Install into your Codex skills directory:

$CODEX_HOME/skills/doctoral-thesis-pipeline

Typical Windows path:

C:\Users\<your-user>\.codex\skills\doctoral-thesis-pipeline

Example prompts:

• Use $doctoral-thesis-pipeline to turn my rough topic into proposal chapters 1-3.
• Use $doctoral-thesis-pipeline to strengthen the logic of my methodology section.
• Use $doctoral-thesis-pipeline to create a literature review workflow and evidence table.
• Use $doctoral-thesis-pipeline to revise my proposal after supervisor feedback.

The included script helps create a seed literature pool when SEMANTIC_SCHOLAR_API_KEY is available.

python scripts/semantic_scholar_search.py --query "AI tourism marketing" --limit 20 --out candidate_papers.jsonl

You can provide the API key through the process environment or .env.local.

• Figures-Diagram for conceptual frameworks, methodology diagrams, taxonomies, timelines, and other non-numeric academic visuals

• Do not fabricate citations, findings, coefficients, or sample statistics
• Prefer narrowing scope over forcing weak completeness
• Align title, framework, variables, and method before expanding prose
• Build evidence before writing from memory on unstable topics
• Produce files, not just commentary, when the user needs thesis work product

MIT