HUAOMP
Holistic, Universal, Absolute, Omni, Meta, Paradigmatic — a six-lens orientation for maximum epistemic breadth when scoping intent and expanding problems before narrowing to action.
HUAOMP is a HUMMBL-aligned stance for intent-to-spec generation and scope expansion. Each lens answers a different question: How does the whole system behave? What is invariant across contexts? What hard boundary cannot be crossed? What does every perspective see? What abstraction level am I missing? What paradigm am I trapped in?
The six lenses
The canonical order is H-U-A-O-M-P. Each lens has a core move, a truth weight, a usefulness weight, and anchor references in the Base120 catalog.
Core move: Trace feedback loops, find leverage points. Look for emergent behavior that no single component explains.
Example: A CI pipeline is slow. The Holistic lens does not ask "which test is slowest?" It asks "where does queueing, feedback delay, and retry amplification create a loop that keeps the team waiting?"
Core move: Extract context-free invariants. What remains true regardless of domain, scale, or implementation?
Example: Every delegation token must be scoped, expiring, and revocable. That invariant holds whether the token is for GitHub, Linear, or a human operator.
Core move: Draw hard boundaries. What is impossible, prohibited, or non-negotiable?
Example: "No agent may commit to
main
without CI green. No agent may bypass the kill switch. No secret
may be hardcoded." These are absolute constraints that shape the
solution space.
Core move: Rotate through all perspectives. What does the operator, the auditor, the customer, and the model see?
Example: A kill switch change looks correct to the engineer who wrote it, but the operator sees a missing runbook, the auditor sees a missing receipt, and the model sees a conflicting instruction.
Core move: Step up one abstraction level. Are we solving the right problem, or optimizing inside a broken frame?
Example: The team is debating which linter to use. The Meta lens asks: "Is the real problem tooling, or is it that no one owns code-quality standards?"
Core move: Name the paradigm, imagine alternatives. What worldview is baked into the question?
Example: "AI agents must be supervised by humans" assumes a human-in-the-loop paradigm. The Paradigmatic lens asks whether a "governance-in-the-loop" paradigm with immutable receipts and circuit breakers could be safer and more scalable.
Using HUAOMP
HUAOMP is used during the earliest stage of problem framing — before a specification, before a plan, before a prompt. A single lens is often enough for a narrow question; a multi-lens combination is used when the stakes are high or the problem is ill-defined.
The combination HA (Holistic + Absolute) is common for
safety reviews: understand the whole system, then draw hard
boundaries. HUAOM is common for strategy: whole-system,
invariant, boundary, perspective, and meta-level views combined. The
full HUAOMP set is reserved for paradigm-shifting
decisions.
Schema
Each HUAOMP lens is defined as a structured object that can be scored and combined programmatically.
| Field | Type | Description |
|---|---|---|
code |
string | Single-letter lens code: H, U, A, O, M, P |
name |
string | Human-readable lens name |
core_move |
string | What the lens does in one sentence |
base120_refs |
tuple[string] | Anchor Base120 model codes |
truth_weight |
float | Lens contribution to truth-seeking |
usefulness_weight |
float | Lens contribution to usefulness |
HUAOMPLens(
code="H",
name="Holistic",
core_move="Trace feedback loops, find leverage points",
base120_refs=("SY1", "RE2", "SY6"),
truth_weight=0.88,
usefulness_weight=0.90,
)Provenance
Origin: HUAOMP was introduced as the HUMMBL epistemic breadth layer — an orientation that sits above the MTSMU truth/utility standard and the Base120 operator catalog.
Implementation: founder_mode/services/huaomp_loop.py — read-only registry and ranking engine for all 63 non-empty lens combinations.
Decision record: The lens definitions and pair bonuses are tested in founder_mode/tests/unit/test_huaomp_loop.py.