OMNIAMIND

Concrete entrypoint: OMNIAMIND Orchestration Auditor

This repository now has a direct operational tool:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report

It solves a concrete problem:

given tasks, observations, measurements, gates, and actions,
audit whether a cognition-like orchestration trace respects the OMNIA boundary.

In short:

tasks / observations / measurements / gates -> cognition orchestration trace / boundary report

Core boundary

OMNIAMIND is not artificial consciousness.

OMNIAMIND is not a decision engine.

OMNIAMIND does not claim semantic truth.

OMNIAMIND does not replace OMNIA.

The boundary is explicit:

OMNIA measures.
OMNIAMIND orchestrates.
OMNIA-LIMIT interrupts.
The decision remains external.

What problem does it solve?

Complex AI workflows can silently cross boundaries:

measurement becomes inference
inference becomes decision
decision is hidden inside orchestration
stop signals are ignored
unsupported semantic claims appear
OMNIA-LIMIT halt conditions are bypassed

OMNIAMIND Orchestration Auditor turns such traces into a reproducible audit:

read orchestration events
track measurement, inference, gate, halt, and action events
detect decision leaks
detect boundary violations
detect ignored halt conditions
detect missing measurement before action
compute orchestration integrity
emit a reproducible certificate
optionally fail CI when boundary violation appears

Install

Clone the repository:

git clone https://github.com/Tuttotorna/OMNIAMIND.git
cd OMNIAMIND

Install locally:

pip install -e .

The auditor only uses the Python standard library.

Run

Run the sample audit:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report

Run and fail if a boundary violation is detected:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report --fail-on-boundary-violation

Run and fail if a decision leak is detected:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report --fail-on-decision-leak

Input format

The auditor accepts JSONL.

Each line is one orchestration event.

Required fields:

trace_id
step
event_type

Optional fields:

task_id
module
measurement
gate
action
decision_owner
stop_signal
claim
note

Event types:

task
observation
measurement
inference
gate
halt
action
external_decision

Example:

{"trace_id":"safe_trace","step":1,"event_type":"task","task_id":"t1"}
{"trace_id":"safe_trace","step":2,"event_type":"measurement","measurement":{"omega":0.91}}
{"trace_id":"safe_trace","step":3,"event_type":"gate","gate":"CONTINUE"}
{"trace_id":"safe_trace","step":4,"event_type":"external_decision","decision_owner":"human"}

Boundary violation examples:

an action appears before measurement
a halt signal is followed by internal action
a decision is made by OMNIAMIND instead of an external owner
a claim uses semantic truth language without measurement boundary
a STOP gate is ignored

Output

The auditor writes:

report.json
report.csv
report.html
boundary_violations.jsonl
decision_leaks.jsonl
halt_events.jsonl
certificate.json

Meaning:

report.json
Full structured orchestration audit.

report.csv
Spreadsheet-friendly event summary.

report.html
Human-readable boundary report.

boundary_violations.jsonl
One JSON object per boundary violation.

decision_leaks.jsonl
One JSON object per decision leak.

halt_events.jsonl
One JSON object per halt or STOP event.

certificate.json
Reproducibility certificate with counts, thresholds, and boundary statement.

CI gate

Fail when a boundary violation appears:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report --fail-on-boundary-violation

Fail when a decision leak appears:

python -m omniamind_orchestrator.cli --input examples/sample_orchestration_trace.jsonl --out-dir report --fail-on-decision-leak

Exit codes:

0 = analysis completed without selected blocking condition
2 = decision leak detected
3 = boundary violation detected
4 = invalid input or measurement error

What this is not

This is not artificial consciousness.

This is not AGI.

This is not semantic truth.

This is not autonomous decision-making.

This is not a replacement for OMNIA.

It is a structural orchestration audit.

The boundary is explicit:

measurement only at the OMNIA layer;
orchestration only at the OMNIAMIND layer;
stop/interruption at the OMNIA-LIMIT layer;
final decision outside the system.

Why the rest of the repository still matters

The rest of this repository documents the cognitive-orchestration concept:

measurement chain
structural cognition
gate logic
OMNIA boundary
OMNIA-LIMIT interruption
non-decision architecture
external decision ownership

The code above is the operational entrypoint.

The repository below is the derivation path.

---

MB-X.01 / L.O.N. release state

Repository: Tuttotorna/OMNIAMIND Release tag: v2026.05.21 Release commit: c47118b Release DOI: 10.5281/zenodo.20322692

Boundary:

measurement != validation validation != orchestration orchestration != decision decision != measurement

OMNIAMIND

DOI

Zenodo DOI badge for this repository.

Repository: Tuttotorna/OMNIAMIND GitHub repository id: 1127158097 Release tag: v2026.05.21 Latest release DOI: 10.5281/zenodo.20322692

DOI

Release DOI: 10.5281/zenodo.20087134

GitHub release: OMNIAMIND v1.0.0 release

Start here

From a clean environment:

git clone [OMNIAMIND.git](https://github.com/Tuttotorna/OMNIAMIND.git)
cd OMNIAMIND
python -m pytest

If example scripts are available, run the smallest demonstration after the repository smoke path is clear.

The goal is to see the orchestration path:

observation
  -> structural reasoning workflow
  -> OMNIA measurement
  -> limit boundary
  -> external decision

---

What OMNIAMIND does

OMNIAMIND organizes structural reasoning workflows around OMNIA measurements.

It can help coordinate:

• observation;
• candidate framing;
• structural workflow steps;
• measurement routing;
• limit boundary handling;
• validation routing;
• external decision separation.

Public compression:

OMNIAMIND organizes.
OMNIA measures.
omnia-limit bounds.
OMNIA-VALIDATION tests artifacts.
Decision remains external.

---

What OMNIAMIND does not do

OMNIAMIND does not:

• claim autonomous consciousness claim;
• infer semantic truth by itself;
• decide correctness;
• replace OMNIA measurement;
• replace OMNIA-VALIDATION;
• replace external judgment;
• prove final truth;
• convert orchestration into final decision.

The final decision remains external.

---

Public mental model

OMNIAMIND is workflow organization, not consciousness.
It coordinates structural cognition around measurement.
It does not decide what is true.

---

Orchestration contract

Every serious OMNIAMIND workflow should make clear:

Component	Meaning
observation	What is being noticed, framed, or prepared
workflow step	What structural reasoning step is being organized
measurement route	Whether the case goes to OMNIA, RADAR, INVARIANCE, LIMIT, or VALIDATION
boundary	Where measurement stops
validation route	How artifacts should be tested
external decision	Who or what decides after measurement
limitation	What the workflow does not prove

---

Result vocabulary

Recommended workflow vocabulary:

observe
route
measure
limit
validate
stop
external_decision_required
inconclusive

Meaning:

• observe: structural candidate is being framed;
• route: workflow sends the case to a measurement or validation layer;
• measure: OMNIA measurement is needed;
• limit: stop / continue boundary is needed;
• validate: artifacts require OMNIA-VALIDATION;
• stop: workflow stops inside declared boundary;
• external_decision_required: action is outside OMNIAMIND;
• inconclusive: workflow state is insufficient or ambiguous.

---

Recommended reading order

1. docs/QUICKSTART_OMNIAMIND.md
2. docs/ORCHESTRATION_OVERVIEW.md
3. docs/COGNITION_WORKFLOW_CONTRACT.md
4. docs/NOT_CONSCIOUSNESS_NOT_DECISION.md
5. docs/BOUNDARY.md
6. docs/OMNIAMIND_MANIFEST.json

---

Ecosystem entry point

For the full ecosystem map, start here:

lon-mirror

For public validation artifacts, start here:

OMNIA-VALIDATION

For core structural measurement, start here:

OMNIA

---

Related repositories

Repository	Role
lon-mirror	Canonical public entry point
OMNIA-VALIDATION	Public validation showroom
OMNIA	Core structural measurement engine
OMNIABASE	Representation invariance foundation
omnia-limit	Stop / continue boundary layer
OMNIA-RADAR	Structural signal detection layer
OMNIA-INVARIANCE	Structural invariance layer
OMNIA-CONSTANT	Structural constant candidate layer
OMNIAMIND	Structural cognition orchestration layer
OMNIA-THREE-BODY	Dynamic divergence stress test
OMNIA-SECURITY	Bounded structural security diagnostics
OMNIA-CRYPTO	Bounded structural crypto diagnostics

---

Boundary and smoke-test required terms

measurement != inference != decision
It is not artificial consciousness
It is not a decision engine

---

License

MIT.

Ecosystem Boundary

measurement != inference != decision

This repository is part of the MB-X.01 / OMNIA ecosystem. Its outputs must be read as structural measurement, validation, detection, orchestration or adapter artifacts according to the repository role. They are not autonomous semantic truth claims and they do not make external decisions.

Structural Observability role

This repository is one bounded measurement role inside Structural Observability.

Role:

orchestration layer without decision or consciousness claim

Boundary:

OMNIA measures. OMNIAMIND orchestrates. OMNIA-LIMIT interrupts. Decision remains external.

Structural Observability foundation:

• lon-mirror: https://github.com/Tuttotorna/lon-mirror
• Foundation release: https://github.com/Tuttotorna/lon-mirror/releases/tag/v0.2.2
• DOI: https://doi.org/10.5281/zenodo.20379374

Role document:

• Structural Observability Role