Agent & Instrument Diagram

How do agents orchestrate a skill into an operating business?

Agents

Agent	Type	Role	Phase
Business Idea Generator	Orchestrator	Coordinates full flow from input to operating venture	All
Domain Researcher	AI Worker	Market analysis, competitor research, friction identification	1
Eval Runner	Quality Gate	LLM-as-Judge validates AI-generated components	1-2
Sprout Generator	Infrastructure	Creates technical substrate when validation gate passes	1
Operations Seeder	Automation	Populates WorkCharts, sales pipeline, marketing cadence	2
Agent Workforce	Delegation	AI agents assigned to WorkChart stages for ongoing operations	3

Instruments

Instrument	Type	Measures	Action
VFL Scorer	Gate	Idea alignment with VVFL criteria	Pass → proceed; Fail → stop or pivot
Optimal Stopping	Gate	Research signal sufficiency (37% rule)	Enough → create venture; Not enough → keep researching
Explore-Exploit	Selector	Channel/ICP/positioning effectiveness	Exploit best; Explore new when stale
SPCL Scoring	Prioritizer	Content authority potential per topic	Highest score → publish first
Compound Rate Tracker	Monitor	Reusable components added back to platform	Growing → healthy; Flat → extractive
Sales Forecasting	Estimator	Deal probability for domain	Realistic → proceed; Unrealistic → adjust

Flow

PERSON WITH A SKILL
       │
       ▼
┌─────────────────────────┐
│  BUSINESS IDEA GENERATOR │ ←── Orchestrator
│  (venture_business_ideas)│
└──────────┬──────────────┘
           │
    ┌──────┴──────┐
    ▼              ▼
DOMAIN         EVAL RUNNER
RESEARCHER     (quality gate on
(AI market      every AI output)
 analysis)          │
    │               │
    └──────┬────────┘
           ▼
   ┌───────────────┐
   │  VFL SCORER   │ ←── Gate Instrument
   │  Is this idea │
   │  virtuous?    │
   └───────┬───────┘
           │
    Pass   │   Fail → Stop
           ▼
   ┌───────────────┐
   │   OPTIMAL     │ ←── Gate Instrument
   │   STOPPING    │
   │   Enough      │
   │   signal?     │
   └───────┬───────┘
           │
    Yes    │   No → Keep researching
           ▼
   ┌───────────────┐
   │    SPROUT     │ ←── Infrastructure Agent
   │   GENERATOR   │
   │  (creates     │
   │   venture)    │
   └───────┬───────┘
           │
           ▼
   ┌───────────────┐
   │  OPERATIONS   │ ←── Automation Agent
   │   SEEDER      │
   │  WorkCharts   │     Instruments:
   │  Sales        │     - SPCL Scoring
   │  Marketing    │     - Explore-Exploit
   │  Accounting   │     - Sales Forecasting
   └───────┬───────┘
           │
           ▼
   ┌───────────────┐
   │    AGENT      │ ←── Delegation Agent
   │   WORKFORCE   │
   │  AI handles   │     Instrument:
   │  admin,       │     - AI Tuning Loop
   │  human        │     - Compound Rate Tracker
   │  handles      │
   │  craft        │
   └───────────────┘

Feedback Loops

Loop	From	To	Signal	Frequency
Algorithm Learning	WorkChart receipts	VFL Scorer + Eval Runner	Did the scored idea produce real value?	Per receipt
Cross-Venture	Venture A receipts	Venture B templates	Which templates work for similar domains?	Per venture
Agent Tuning	Agent performance data	AI Tuning Loop	Which agent configs produce better outcomes?	Weekly
Compound Rate	Reusable components	Platform template library	Is this venture contributing back?	Monthly
Founder Feedback	Human judgment	ICP + Value Prop refinement	Did the generated business match reality?	Per milestone

The Meta-Algorithm

The venture creation flow IS the Universal Algorithm:

INTENT (person has a skill + sees a gap)
  → ROUTE (algorithms score idea, find best model and market)
  → INFRASTRUCTURE (Sprout Generator builds operating substrate)
  → SETTLE (venture launches, first WorkCharts execute)
  → FEEDBACK (outcomes feed algorithms, improving next venture)

Every venture that runs on the mycelium makes the next venture better. The algorithms learn from aggregate WorkChart receipts across all ventures.

Key Finding

The acceptance test for the entire pipeline is the first feedback loop completing: a real founder inputs a real skill → the generator produces a real venture → the venture executes real WorkCharts → the receipts improve the generator for the next founder. Until that loop closes, every component is an assumption. The VFL Scorer and Optimal Stopping Gate are the two critical instruments — they decide what gets built and when. If those two instruments work, the rest is plumbing.

Context

Capability Map — Previous: what can we actually do?
Pictures Index — Hub: all five maps
A&ID Template — Nomenclature reference
A2A Sprout A&ID — Infrastructure layer A&ID

Agents​

Instruments​

Flow​

Feedback Loops​

The Meta-Algorithm​

Key Finding​

Context​