Plan

Plan

When to Use

• After topic and chapter outline are settled and the user is ready to plan execution
• User mentions a submission deadline, defense date, or university requirement
• Existing plan needs updating due to changed circumstances
• User asks "when should I finish X?" or "how do I fit this in before my deadline?"

Procedure

Step 0: STOP AND ASK

Before generating any output, you MUST confirm the following. If the answers are already clear from the user's triggering message, verify with a single confirmation question. If any are unclear, ASK BEFORE PROCEEDING. Do not embed these as "clarifying questions" in the output document — ask them conversationally first.

Mandatory confirmations:

• Specific defense date (not just month — e.g., "June 15, 2027", not "June 2027")
• Submission deadline (often 2–8 weeks before defense at Chinese universities — do not assume)
• Any intermediate milestones: proposal defense (开题报告 — required at most Chinese universities before empirical work), advisor review deadlines, committee draft submissions
• Current chapter status (cross-check with progress.md if it exists)

Only proceed to the next steps once these are confirmed.

1. Gather inputs

Read the following if they exist:

• meta/progress.md — current chapter status and completion levels
• meta/milestones.md — any existing milestones already recorded
• meta/plan.md — any existing plan to update rather than replace

Then ask the user for any missing information:

1. Submission deadline — What is the university submission date? (exact date)
2. Defense date — Is there a defense/viva date? (if applicable)
3. Chapter count — How many chapters are planned? What are they?
4. Chapter status — Which chapters are done, in progress, or not started? (cross-check with progress.md if it exists)
5. Weekly capacity — How many hours per week can you realistically dedicate?
6. Buffer preference — How much buffer time do you want before the submission deadline?

2. Work backward from the submission deadline

Use backward planning from the submission deadline:

1. Reserve buffer time before the deadline (suggest 1–2 weeks minimum)
2. Reserve time for final formatting, proofreading, and export (suggest 1 week)
3. Reserve time for final review pass across all chapters (suggest 1–2 weeks)
4. Allocate remaining time to chapters based on their current status:
   • [TODO] chapters need the most time (drafting + review)
   • [OUTLINED] chapters need less (drafting + review)
   • [DRAFTING] chapters need moderate time (finish draft + review)
   • [REVIEWING] chapters need the least (finish review cycle)
5. Set milestones for each chapter: outline complete, first draft, reviewed
6. Set weekly goals that are achievable given the user's stated capacity

3. Check for realism

Flag if the plan is infeasible:

• If there is not enough time to complete all chapters before the deadline, say so clearly
• Suggest what to prioritize if time is tight (core chapters first)
• Suggest what could be cut or condensed if needed
• Recommend a minimum viable dissertation path if deadlines are very tight

4. Write meta/plan.md

Output the plan to meta/plan.md using this structure:

# Dissertation Plan

## Deadlines
- Submission: [date]
- Defense: [date, if applicable]
- Advisor review cutoff: [date — buffer before submission]

## Phases

### Phase 1: [Name] — [date range]
Goal: [What should be achieved]
Chapters: [Which chapters to focus on]
Milestone: [Concrete deliverable]

### Phase 2: [Name] — [date range]
...

## Weekly Goals

### Week of [date]
- [ ] [Specific, achievable goal]
- [ ] [Specific, achievable goal]

### Week of [date]
...

## Buffer Time
- [Start date] – [submission date]: Final review, formatting, proofreading

## Assumptions
- Weekly writing capacity: [N] hours
- [Any other assumptions made]

## Risks
- [Any identified risks or tight spots in the timeline]

5. Update meta/milestones.md

If meta/milestones.md exists, update it with the milestone dates from the plan. If it does not exist, create it with the key milestones identified.

6. Summarize and confirm

Present a brief summary of the plan to the user:

• Key dates and phases
• Most critical near-term weekly goals
• Any risks or tight spots
• Ask if they want to adjust anything

Inputs

• meta/progress.md — chapter completion status
• meta/milestones.md — existing milestone tracking (if present)
• meta/plan.md — existing plan to update (if present)
• User-provided: submission deadline, defense date, chapter count, weekly capacity

Outputs

• meta/plan.md — timeline with phases, milestones, weekly goals, and buffer time
• Updated meta/milestones.md — milestone dates

Notes

• Always work backward from the deadline — this is the most reliable planning method
• Be honest about infeasibility. A tight plan is useless; a realistic one is actionable.
• Weekly goals should be concrete and small enough to actually complete in a week
• Buffer time is non-negotiable — always include it before the submission deadline
• Revisit the plan when progress.md shows chapters are behind or ahead of schedule
• University submission dates are hard deadlines; defense dates may have more flexibility
• If the user does not know their deadlines yet, help them estimate based on typical dissertation timelines

Chinese university context: Most Chinese PhD programs require a proposal defense (开题报告 / kaiti baogao) before significant empirical work. This typically happens 3–6 months into the PhD and must be on the timeline. Also common: midterm evaluation (中期考核), pre-defense review (预答辩), and formal defense (正式答辩). For international/US programs, these may be called: prospectus defense, qualifying exams, committee meetings, defense.