Changelog - 2025-11-27 (#1)

Emerging Manager Outline with Proprietary Scorecard Vocabulary

Overview

Created a comprehensive outline for evaluating LP commitments to emerging manager VC funds, integrating Hypernova Capital's proprietary 12-dimension GP evaluation framework directly into the memo generation system.

New File: templates/outlines/lpcommit-emerging-manager.yaml

Size: 1,806 lines

Purpose

Streamlined structure for evaluating LP commitments to first-time fund managers, solo GPs, and funds under $50M where institutional track record data may be limited.

Key Innovation

Unlike generic templates, this outline embeds proprietary vocabulary and evaluation frameworks directly into the guiding questions, preferred sources, and section requirements—ensuring AI-generated memos reflect the firm's actual thinking, not generic analysis.

The 12-Dimension Scorecard Framework

Three Dimension Groups (Strategically Placed)

Each Dimension Includes

For each of the 12 dimensions, the outline provides:

1. Definition - Full description from the Hypernova scorecard

2. Evaluation Questions - 4 specific questions to research

3. Evidence Sources - Where to find supporting data

4. Red Flags - Warning signs to watch for

5. Scoring Rubric - 1-5 scale with detailed criteria

Example: "Hustle" Dimension

- term: "hustle"
  usage: "Super-human level of omnipresence and energy"
  definition: |
    Rising above the world of existing early-stage venture firms requires
    sweat-inducing levels of energy. The methods are myriad, but the result
    is a super-human level of omnipresence.
  evaluation_questions:
    - "What is the GP's meeting cadence and founder outreach volume?"
    - "How quickly do they respond to opportunities?"
    - "Are they consistently present at key ecosystem events?"
    - "Do they create their own deal flow through proactive outreach?"
  evidence_sources:
    - "Meeting and outreach metrics"
    - "Event attendance and hosting"
    - "Response time to opportunities"
  red_flags:
    - "Passive investment approach"
    - "Slow response to opportunities"
    - "Limited proactive founder engagement"
  scoring:
    5: "Exceptional energy; omnipresent in ecosystem; creates own luck"
    4: "High energy; proactive approach; consistently present"
    3: "Adequate activity level; standard VC engagement"
    2: "Below average activity; reactive approach"
    1: "Passive; minimal proactive engagement"

Proprietary Vocabulary Terms

Beyond the 12 scorecard dimensions, the outline includes proprietary terminology:

Section Structure (10 Sections)

1. Executive Summary - Investment thesis, recommendation, key risks

2. GP Background & Credibility - Character, jagged resume, theory of market (Group 1 dimensions)

3. Fund Strategy & Thesis - Investment focus, right to win, dealflow (Group 2 dimensions)

4. Portfolio Construction - Check sizes, reserves, concentration

5. Value-Add & Differentiation - Support model, signal power

6. Early Signals & Leading Indicators - Bold bets, early breakouts (Group 3 dimensions)

7. Fee Structure & Economics - Management fees, carry, alignment

8. LP Base & References - Co-investors, founder references

9. Risks & Mitigations - Key person risk, market risk

10. Recommendation - Final assessment with conviction level

Additional Features

Custom Memo Date Support

Added memo_date field to company data JSON:

{
  "name": "Don't Quit Ventures",
  "type": "fund",
  "mode": "justify",
  "outline": "lpcommit-emerging-manager",
  "memo_date": "January 10, 2025",
  ...
}

• Export now uses custom date instead of today's date

• Supports multiple formats: YYYY-MM-DD, Month DD, YYYY, etc.

• Falls back to today's date if not specified

Citation Consolidation Utility

New cli/utils/consolidate_citations.py:

• Consolidates scattered per-section citation blocks into ONE block at document end

• Globally renumbers footnotes sequentially

• Fixes malformed markdown concatenation issues

Files Changed

New Files (3)

Modified Files (8)

Why This Matters

Problem Solved

AI-generated memos risk being too generic. Without firm-specific frameworks, analysis lacks differentiation and doesn't reflect the firm's actual investment thesis.

Solution

By embedding proprietary vocabulary, evaluation questions, and scoring rubrics directly into the outline:

1. Research agents know what evidence to prioritize

2. Writer agents use firm-specific terminology

3. Output reflects the firm's actual thinking, not generic AI content

4. Scorecards can eventually be auto-generated from memo content

Usage

# Create company data file with outline reference
cat > data/MyFund.json << 'EOF'
{
  "name": "My Fund",
  "type": "fund",
  "mode": "consider",
  "outline": "lpcommit-emerging-manager",
  "memo_date": "January 15, 2025"
}
EOF

# Generate memo
python -m src.main "MyFund"

# Export with custom date
python cli/export_branded.py output/MyFund-v0.0.1/4-final-draft.md --brand hypernova --mode dark
# Output: "Using custom memo date: January 15, 2025"

Related Documentation

• templates/scorecards/lp-commits_emerging-managers/hypernova-scorecard.md - Source scorecard

• context-vigilance/Model-Scorecard-Agent-and-Template-System.md - Scorecard agent architecture (planned)

Next Steps

• Implement Scorecard Agent to auto-score dimensions from memo content

• Add scorecard table insertion to memo sections

• Create direct investment scorecard template

Contributors

Implementation by Claude Code (Anthropic) with user guidance.