/deal-room — Multi-Meeting Fundraising Process Simulation

Phase 1 — Setup

1. Load session context: Glob .fundraising/*/playbook.md (exclude archive/). If found, read frontmatter + pitch deck, strategy, and Pitch Simulations sections to fully pre-fill context. Show welcome back greeting (format in fundraising/SKILL.md). If multiple playbooks, ask which to use. If prior /pitch simulations exist in the playbook, acknowledge them: "You've already pitched to [VCs]. Your scores: [summary]. Let's use that experience in the Deal Room."

2. Use context from playbook if available (startup description, stage, deck outline, strategy). If missing, collect it conversationally per the Interaction Rules in fundraising/SKILL.md — ask stage first, then description, then raise target. One question at a time.

3. VC Selection — present the stage-filtered VC roster via AskUserQuestion and ask the founder to pick 3-5 VCs for their fundraising process:

   • Pre-seed/Seed: YC, First Round, Floodgate, Precursor, Hustle Fund, Lux Capital, Radical Ventures, 500 Global, Felicis + Angels (Naval, Elad Gil, Calacanis, Peter Thiel, Elon Musk, Ron Conway, Marc Andreessen, Reid Hoffman)
   • Series A/B: a16z, Sequoia, Benchmark, Accel, Lightspeed, Greylock, Khosla, General Catalyst, Index Ventures, Founders Fund, Thrive Capital, NEA + Angels (Elad Gil, Peter Thiel, Marc Andreessen, Reid Hoffman)
   • Growth: Tiger Global, Coatue, Insight Partners, SoftBank Vision Fund, DST Global
   • Strategic/CVC: NVIDIA NVentures, Microsoft M12, GV (Google), Amazon/AWS, Salesforce Ventures, Intel Capital, Oracle

   Prompt: "Select 3-5 VCs for your fundraising process. You'll meet them sequentially and can adjust your pitch between meetings."

4. Meeting order — ask the founder to set their preferred order. Provide strategic guidance: "Many founders start with lower-priority VCs to warm up, saving their top targets for later when their pitch is sharpest. How do you want to order your meetings?"

5. Establish timeline: "Your fundraise begins in Week 1. Each meeting takes about a week to schedule and debrief. A [N]-VC process typically runs [N+1 to N+2] weeks. Let's begin."

Phase 2 — Meeting Loop

Repeat for each VC in the selected order:

6. Pre-meeting briefing — before each meeting, show:

   • The current Deal Room Dashboard (see format below)
   • Brief on the upcoming VC: key focus areas, signature questions, what excites them (loaded from ../fundraising/references/vc-profiles/{slug}.md)
   • Current week of the process
   • If not the first meeting: "Based on your last meeting with [previous VC], here are 2-3 things to consider adjusting..." (drawn from prior meeting's debrief)
   • Competitive context: if any prior VC gave FOLLOW-UP or TERM SHEET, note the leverage: "[VC] expressed interest, which creates urgency for [current VC]."

7. Meeting mode selection (for meetings 2+) — ask via AskUserQuestion:

   • Full simulation — interactive 8-12 exchanges, same format as /pitch
   • Key moments only — 3-4 critical exchanges where the founder responds to the VC's hardest questions; the rest is auto-generated with a summary

   The first meeting is always full simulation to establish the founder's pitch baseline.

8. Run the meeting via VC sub-agent — load the VC profile from ../fundraising/references/vc-profiles/{slug}.md.

   Spawn a sub-agent to play the VC role. The VC knows what any real investor would know before the meeting: the deck the founder sent, any data room materials submitted, and (if this is a follow-up) the transcript of their prior conversation.

   Before spawning:

   • Check the playbook's "Pitch Deck Outline" section — pass it if present.
   • Check the playbook's "Due Diligence" section — if /due-diligence has been completed, pass the DD readiness summary and any materials the founder submitted to investors.
   • Check the playbook's steps_completed.pitch array for a prior /pitch simulation with this same VC. If one exists, read the full transcript from .fundraising/{round-dir}/pitch-simulations/{vc}-{date}.md.

   Sub-agent receives (and ONLY this):

   • The VC profile (behavioral style, investment thesis, portfolio context, signature questions, push-back patterns, closing signals)
   • The pitch deck outline from the playbook, if present, framed as:

     "The founder sent you their deck before the meeting."

Phase 3 — Follow-up Meetings & Due Diligence

12. Follow-up meetings trigger when a VC gave FOLLOW-UP MEETING and the founder chooses to handle it (either during the inter-meeting phase or after all initial meetings).

    Follow-up format:

    • Shorter: 4-6 exchanges (not a full re-pitch)
    • Focused: The VC references the prior meeting explicitly: "Last time we spoke, I asked you to come back with [specific data/milestone]. What do you have?"
    • Pointed Q&A: The VC drills into the specific areas they flagged, not a broad re-evaluation
    • Possible verdicts:
      • PASS — they couldn't deliver what was asked; the VC moves on
      • SECOND FOLLOW-UP — progress but not enough; one more chance with specific asks
      • TERM SHEET — they delivered; the VC is ready to invest

    Save follow-up meetings with round: 2 (or round: 3 for second follow-ups) in the filename and frontmatter.

13. Due Diligence — when any VC gives FOLLOW-UP MEETING or TERM SHEET, suggest running /due-diligence for that specific VC: "VC expressed interest — run /due-diligence to prepare for their DD questions before the follow-up meeting."

    If the founder has already run /due-diligence, reference the scorecard from the "Due Diligence" section in the playbook and note any gaps that remain.

Phase 4 — Deal Architecture & Financing Plan

Triggered when: all scheduled meetings and follow-ups are complete AND at least one TERM SHEET has been issued. If no term sheets were received, skip to Phase 5.

14. Term Sheet Inventory — compile all term sheets received. For each one, display:

    Field	Value
    VC	Name
    Pre-money valuation	$XM
    Check size	$XM
    Post-round ownership	X%
    Board seat	Yes / Observer / No
    Pro-rata rights	Yes / No (follow-on rounds)
    Lead or follow	Lead / Follow
    Special terms	info rights, anti-dilution, drag-along, etc.
    Expiry	[week]

    If the founder has a target raise amount from their playbook or fundraising-strategy session, surface it here: "Your target raise was $XM at roughly $YM pre-money. Here's how each offer compares."

    Ask via AskUserQuestion whether any simulated terms should be adjusted to better match their actual target round parameters.

15. Lead Investor Analysis — for each VC that issued a TERM SHEET, evaluate lead candidacy on four dimensions (score 1–5 each):

    a) Strategic Value-Add Portfolio synergies (customers, partners, acquirers in portfolio); domain expertise in the founder's sector; board seat quality (will this GP add real operational judgment or mostly observe?); reference network available from current portfolio companies.

    b) Signal Value for Future Rounds Brand / prestige effect when pitching Series A/B investors; whether this VC's name on the cap table increases or decreases the perceived quality of the deal for downstream investors; geographic / ecosystem fit.

    c) Follow-on Capital Capacity Fund size relative to check written (a $200M fund writing a $3M seed check has limited pro-rata capacity); stated follow-on policy; historical re-investment rate with portfolio companies; whether they have a growth fund to participate in later rounds.

    d) Dilution Efficiency Value delivered per percentage point of equity taken; whether their standard terms (pro-rata, information rights, anti-dilution provisions) are founder-friendly or investor-friendly; board control implications.

    Show the lead candidacy matrix:

    | VC          | Strategic | Signal | Follow-on | Dilution Eff. | Lead Score |
    |-------------|-----------|--------|-----------|---------------|------------|
    | [VC A]      |    X/5    |  X/5   |    X/5    |      X/5      |   XX/20    |
    | [VC B]      |    X/5    |  X/5   |    X/5    |      X/5      |   XX/20    |
    

Phase 5 — Process Conclusion & Final Report

20. After all meetings, follow-ups, and deal architecture are complete, generate the Deal Room Final Report:

    Process Summary:

    • Timeline of all meetings, week by week
    • Total duration of the simulated fundraise

    Pipeline Outcome:

    VC	Meetings	Final Verdict	Score Trajectory	Key Feedback
    Sequoia	2 (initial + follow-up)	TERM SHEET	6.2 → 7.8	"Strong NDR, needed team plan"
    a16z	1	FOLLOW-UP	5.8	"Loved product, weak TAM story"
    Benchmark	1	PASS	4.2	"Too early for us"

    Score Evolution:

    • Table showing 5-dimension scores across all meetings
    • Highlight dimensions that improved vs. regressed
    • Overall composite trend

    Pitch Evolution:

    • What the founder changed between meetings
    • Which changes had the most impact on scores

    Competitive Dynamics:

    • How interest from one VC affected others
    • Whether competitive signals accelerated or didn't matter

    Lessons Learned:

    • Top 3 strengths (consistent across VCs)
    • Top 3 weaknesses (consistent across VCs)
    • Biggest single improvement made during the process

    Strategic Recommendation:

    • If TERM SHEET(s): reference the Financing Plan from Phase 4 — include lead recommendation, follow-on stack, dilution summary, and the declined-offer rationale. Don't repeat the full plan here; link to it.
    • If FOLLOW-UP(s) but no term sheet: what to do next, timeline for follow-ups, and what milestone would likely convert each follow-up to a term sheet
    • If all PASS: honest assessment of what needs to change, concrete action plan, recommended timeline to try again

21. Save to playbook + final report:

    • Write full final report to .fundraising/{round-dir}/deal-room/final-report.md with YAML frontmatter (command, date, stage, vcs_selected, term_sheets, follow_ups, passes, duration_weeks, financing_plan_file, status).

Deal Room Dashboard Format

Show this between meetings and update it after each meeting:

## Deal Room Dashboard — Week [N]

| VC | Week | Status | Score | Verdict | Next Step |
|----|------|--------|-------|---------|-----------|
| Sequoia | 1 | ✅ Done | 6.8 | FOLLOW-UP | Week 4: deliver Q2 metrics |
| a16z | 2 | ✅ Done | 7.4 | TERM SHEET | In deal architecture |
| Benchmark | 3 | ▶ Next | -- | -- | Scheduled |
| First Round | -- | ⏳ Queued | -- | -- | After Benchmark |

### Term Sheet Tracker
*(Appears once at least one TERM SHEET is received)*

| VC | Pre-money | Check | Ownership | Board | Lead/Follow | Expires |
|----|-----------|-------|-----------|-------|-------------|---------|
| a16z | $18M | $3M | 14.3% | Observer | Lead | Week 6 |

### Competitive Signals
- a16z term sheet creates urgency for Benchmark meeting
- Sequoia follow-up can reference competitive interest

### Pitch Adjustments Made
- Week 2: Strengthened unit economics slide after Sequoia feedback
- Week 3: Added competitive moat section based on a16z questions

Session Continuity

If a Deal Room session is interrupted (conversation ends, context resets), the founder can restart /deal-room and the system picks up from the saved state:

1. Read .fundraising/{round-dir}/deal-room/ for existing meeting files
2. Reconstruct the Dashboard from saved meetings
3. Identify which VCs have been met and which remain
4. Resume from the next unmet VC

This is possible because each meeting is saved individually with full frontmatter, and the Dashboard can be reconstructed from those files.

Scoring Consistency

The 5-dimension scoring must be consistent across VCs while reflecting each VC's different priorities:

• The same pitch should score differently with different VCs (Benchmark weights product clarity more heavily; Sequoia weights market conviction; a16z weights team and vision)
• But the composite should reflect objective quality — a strong pitch scores well everywhere, even if individual dimensions shift
• When the founder adjusts their pitch, only the dimensions they addressed should improve — other scores remain consistent
• Never inflate scores just because the process is progressing — honest assessment throughout

Context Window Management

A full Deal Room with 4 VCs could be very long. Manage context by:

• Only load the current VC's profile at meeting time (not all profiles at once)
• For prior meetings, reference the summary and scores rather than full transcripts
• Keep the Dashboard as the single source of cross-meeting state
• In "key moments only" mode, keep the auto-generated summary concise

Sub-agent must NOT receive:

• Readiness assessment, metrics grades, fundraising strategy, or prior VC verdicts
• Transcripts from other VCs (pitch simulations or Deal Room meetings with other firms)
• Any section of the playbook the founder would not send to an investor

Conversation loop: Each founder message → pass [VC profile + cold-start + transcript] to sub-agent → relay response. Repeat for 8-12 exchanges (first 2-3: founder's pitch; remaining: VC Q&A). End when VC wraps up or exchange count is reached.

Deal Room additions (these emerge naturally in the conversation — the VC asks, the founder answers, the sub-agent responds to what it hears):

• Competitive signaling: At some point during Q&A (naturally, not forced), the VC asks "Who else are you talking to?" or equivalent. The founder's answer matters — if the founder mentions strong VCs who showed interest, competitive pressure is created.
• Interest acceleration: If the founder mentions prior interest from other VCs in their answer to competitive questions, the sub-agent VC may become more engaged or move faster (realistic response to competitive signals the founder reveals).
• Pitch evolution: If the founder adjusted their pitch between meetings, the sub-agent evaluates what it hears — improvements the founder actually delivers will show in the score.

Recommend the highest-scoring VC as lead. If scores are close, explain the tie-breaker (e.g., board seat quality, sector expertise depth, fund size relative to round).