Agent-Skills.md

Agent Skills: Deal Scoring Engine

Automated deal scoring based on thesis alignment, market size, team, and traction metrics

UncategorizedID: a5c-ai/babysitter/deal-scoring-engine

Repository

a5c-ai/babysitter
a5c-aiLicense: MIT
244

Install this agent skill to your local

pnpm dlx add-skill https://github.com/a5c-ai/babysitter/tree/HEAD/plugins/babysitter/skills/babysit/process/specializations/domains/business/venture-capital/skills/deal-scoring-engine

Skill Files

Browse the full folder contents for deal-scoring-engine.

Download Skill

Loading file tree…

plugins/babysitter/skills/babysit/process/specializations/domains/business/venture-capital/skills/deal-scoring-engine/SKILL.md

Skill Metadata

Name
deal-scoring-engine
Description
Automated deal scoring based on thesis alignment, market size, team, and traction metrics

Deal Scoring Engine

Overview

The Deal Scoring Engine skill provides automated, consistent evaluation of investment opportunities against defined criteria. It generates composite scores based on thesis alignment, market opportunity, team quality, and business traction to support pipeline prioritization and investment decisions.

Capabilities

Thesis Alignment Scoring

  • Match opportunities against fund investment thesis
  • Sector, stage, and geography fit assessment
  • Strategic priority alignment scoring
  • Anti-thesis and exclusion criteria flagging

Market Opportunity Assessment

  • TAM/SAM/SOM scoring based on market data
  • Market growth rate and timing assessment
  • Competitive intensity evaluation
  • Regulatory and macro environment scoring

Team Evaluation Scoring

  • Founder background and experience assessment
  • Domain expertise and market knowledge scoring
  • Team completeness and capability gaps
  • Track record and references scoring

Traction and Metrics Scoring

  • Revenue and growth rate benchmarking
  • Unit economics (LTV/CAC, margins) scoring
  • Engagement and retention metrics assessment
  • Capital efficiency and burn rate evaluation

Composite Score Generation

  • Weighted composite scoring with configurable weights
  • Stage-appropriate scoring models (seed vs. growth)
  • Sector-specific scoring adjustments
  • Historical score calibration against outcomes

Usage

Score New Deal

Input: Company data, metrics, team information
Process: Apply scoring models across dimensions
Output: Composite score, dimension scores, flags, recommendations

Configure Scoring Model

Input: Scoring criteria, weights, thresholds
Process: Update scoring model parameters
Output: Configured scoring model, validation results

Benchmark Against Portfolio

Input: Deal scores, portfolio company scores
Process: Compare against portfolio at similar stage
Output: Relative ranking, percentile position, comparisons

Calibrate Model

Input: Historical deals and outcomes
Process: Analyze predictive accuracy, adjust weights
Output: Calibration report, recommended adjustments

Scoring Dimensions

| Dimension | Weight Range | Key Factors | |-----------|--------------|-------------| | Thesis Fit | 15-25% | Sector, stage, geography, strategy | | Market | 20-30% | TAM, growth, competition, timing | | Team | 25-35% | Experience, domain, completeness | | Traction | 20-30% | Revenue, growth, unit economics |

Integration Points

  • Deal Flow Tracker: Embed scores in pipeline management
  • Proactive Deal Sourcing: Score for outreach prioritization
  • IC Memo Generator: Include scores in investment memos
  • Market Sizer: Feed market data into scoring

Best Practices

  1. Calibrate scoring models quarterly against outcomes
  2. Use stage-appropriate models (early vs. late stage)
  3. Document override decisions when departing from scores
  4. Maintain transparency on scoring methodology
  5. Avoid over-reliance on scores for complex decisions