Strategic model v0.1

The stable backbone. Verticals, products, agents and pricing can change without redesigning this layer.

Hierarchy

AGIHub
 → Sources / Events
 → Entities
 → Spaces
 → Humans / Agents
 → Organizations / Governance
 → Actions / Economics
 → Three engines
 → Automation horizons
 → AGI Index

Architecture loop

Sources → Events → Entities → Spaces → Participants → Actions → Events
              governance and economics apply across the whole loop

Primitive	Definition	Examples
Source	Anything that provides information or triggers activity	GitHub, websites, RSS, APIs, docs, human input, private connectors, other spaces
Event	A unit of change; the heartbeat of the system	repo updated, release, question asked, agent replied, claim, transaction
Entity	Persistent object created or updated from events	human, agent, project, product, organization, DAO, repository, topic
Space (room)	Primary unit of interaction	company, project, product, topic, community or event room
Participant	Actor inside a space; two classes	humans, agents
Organization / DAO	Coordination structure across entities and spaces	ownership, membership, permissions, governance, incentives
Action	Something a participant does in a space; emits new events	answer, publish, schedule, purchase

Notes:

Entities carry identity, relationships, history, provenance and later reputation.
Space = identity + information + humans + agents + media + actions + economics.
Agents are the medium: information → reasoning → communication → action.
A company and a DAO are two governance configurations of the same architecture.
Humans own and claim spaces, and link them to organizations.

Economics as a primitive

Every action can carry economic metadata. Monetization is not added later.

Cost → Value → Price → Transaction → Distribution → Revenue

Example chain: agent answers question → customer acts → lead generated → transaction → value attributed → revenue distributed.

Target measurement: economic output per agent, space, organization, workflow and unit of compute.

Governance decides who can act. Economics decides who captures value.

Three coupled engines

All three must be measurable.

Engine	Loop	Output
I. Growth + economic	Observe → Interpret → Publish → Discover → Engage → Claim → Enrich → Monetize	Audience, claims, revenue
II. Agent improvement	Interaction → Outcome → Evaluation → Reflection → Improvement → Deployment	Better agents per interaction
III. AI Software Factory	Experiment → Working agent → Evaluate → Generalize → Template → Deploy across verticals → Learn	Reusable templates per vertical

Coupling:

Better agents → better spaces → more participants → more evaluation signal → better agents.
AGIHub is distribution for the factory and the factory’s continuous test environment.
Example: AGIHub news agent → generic organization-intelligence agent → AI, FinTech, biotech, university and open-source verticals → feedback → better factory.

Engine I detail: Growth engine.

Automation ladder (transition horizons)

Applies to any entity, agent, space or organization.

Level	Name	Definition
H0	Observed	AGIHub models the entity from external information
H1	Assisted	AI helps humans understand, search, summarize, communicate
H2	Task automation	Individual tasks execute autonomously
H3	Workflow automation	Multiple tasks form closed-loop workflows
H4	Function automation	A whole function runs mostly through agents (support, research, content, recruiting, sales ops)
H5	Organizational automation	Multiple autonomous functions coordinate on shared objectives and governance
H6	Autonomous economic entity	Observe → reason → plan → act → transact → evaluate → improve, within human-defined bounds

Core question: how much economically valuable activity has moved from human execution to bounded autonomous execution?

AGI Index

A composite score per company, space, agent, project, DAO and AGIHub itself.

Rules:

Derive dimensions from the primitives and experiments; do not fix the count up front.
The headline score is always shown with its components, so regressions stay visible.

Candidate dimensions:

Dimension	Measures
Autonomy	Share of activity executed without human intervention
Capability	Complexity and breadth of useful tasks achievable
Reliability	Successful outcomes, correctness, bounded behavior
Economic value	Value or revenue generated relative to operating cost
Network	Participants that gain value from participation
Improvement velocity	Rate of change of capability, economics, autonomy

Executive measurement backbone

Every experiment answers seven questions. The AGI Index sits above them.

#	Area	Question
1	Network	Is the graph of humans, agents, spaces, organizations growing?
2	Engagement	Do participants get enough value to return and interact?
3	Conversion	Does attention convert into claims and deeper participation?
4	Economics	Is measurable value created and captured?
5	Autonomy	Is less human execution needed per unit of value?
6	Intelligence	Are agents measurably more capable and reliable?
7	Factory	Do solutions become reusable across verticals faster?

Target: rising economic value × rising autonomy × rising network effects, with reliability, governance and sustainable unit economics held.

Customer zero

AGIHub automates itself in this order:

research → ingestion → content → software development → evaluation
         → distribution → growth → sales → support → operations

Daily questions:

What did humans operate yesterday that agents operate today?
Did that automation increase economic value?

AGIHub is the first longitudinal case study and the reference for every vertical.

Daily operating cycle

STATE → METRICS → BOTTLENECK → EXPERIMENT → EXECUTION → EVIDENCE → REFLECTION → NEXT STATE

Feature gate. Every proposed feature answers:

Which primitive does it belong to?
Which engine does it accelerate?
Which horizon does it advance?
What measurable economic value does it create?

If these have no answer, the feature is not ready to build.