Original AI-generated sigil · Seven nodes for the proposed editions.
Three Fields, Three Jobs
Version, Edition & Rating
Version
Identifies a particular release of a component or package. Existing component versions stay intact until their definitions or implementation change.
AkashicNEXUS v0.1.0
Edition
Names the proposed maturity band and its visual identity. Advancement requires evidence, rather than a colour change.
Bronze Edition
Rating
A scoped assessment against the 1–10 maturity rubric, with its basis, limitations and next evidence recorded.
Provisional 6/10
The workflow diagram keeps its separate v0.2.0 identifier. Package releases, framework releases, content editions and BQ evidence levels are distinct fields. There is no automatic equivalence between v0.6.0 and Level 6.
Seven Named Editions
From Foundations To A Living Ecosystem
This is a proposed maturity framework adopted for this reference edition. Stone, Iron and Copper are retrospective judgements. Bronze is provisional. No formal historical pass dates have been invented.
Current NEXUS position. The public guide and private coordination workspace exist; systematic end-to-end validation is still pending.
Silver
7/10
Verified End-To-End Workflow
Representative tasks complete from input to checked delivery. Changed inputs produce a traceable rerun and the previous output remains available.
Not Yet Demonstrated
Evidence & Advancement
Required Evidence
Observed end-to-end runs, delivery receipts, failure handling and a changed-input rerun.
NEXUS Assessment Basis
Target criteria only. No pass date is assigned.
Gold
8/10
Reliable Connected Execution
Tools exchange structured outputs reliably. Dependency changes identify stale outputs and trigger controlled reruns, with effective HOLD and access controls.
Automatic execution across the full toolbox has not been established.
Platinum
9–10/10
Validated Living Ecosystem
Adaptive routing is independently compared and sustained operation is demonstrated across varied workflows. Recovery, oversight and reproducibility remain explicit.
Not Yet Demonstrated
Evidence & Advancement
Required Evidence
Independent evaluations at Level 9; sustained reproducibility, reliability and recoverable failures at Level 10.
NEXUS Assessment Basis
An aspiration with evidence requirements, not a claim of perfection.
A Scoped Design Assessment
Component Maturity Ratings
These preserve the provisional ratings discussed during design. They are judgements from the definitions and implementations reviewed, without independent scoring. They measure maturity, rather than truth, spiritual attainment or the quality of a person.
AkashicMESSAGE
Communication & Attribution Lens
6/10
v0.4.0-beta.1
Why This Rating?
Current Basis
Structured claim separation and attribution guidance is available.
What Would Raise It?
Demonstrate consistent attribution across varied messages with independent checks.
Meta: Not Assessed · Mettā: Not Assessed
Overall maturity is not reused as either mode score. No independent validation is claimed.
Adaptive Orchestration. Routing and outcomes have independent comparative evaluation.
Proven Living Ecosystem. Sustained, reproducible operation across varied tasks and recoverable failures is demonstrated.
The earlier overall 6/10 is an integrated design judgement, not the arithmetic mean of these component scores. Visual presentation, analytical reliability and automated execution must be checked separately.
AkashicSTATE · v0.1.0-draft.1
The Metadata & Maturity Module
AkashicASSESS makes an assessment. AkashicSTATE preserves the result, its scope and its evidence. It records version, edition, maturity, validation, dependencies, visibility, HOLD, recommendations and review history.
Meta Mode
Examines definition quality, versions, evidence, dependencies and reproducibility. Applying AkashicSTATE to itself is a meta operation.
Current Separate Score: Not Assessed
Mettā Mode
Examines care, agency, cultural attribution, accessibility and respectful uncertainty. It is a practice and review lens across every layer.
Current Separate Score: Not Assessed
Meta and Mettā are modes, not extra editions or higher ranks. Scores stay separate and require evidence. Missing evidence is recorded as Not Assessed, rather than zero. Irrelevant criteria need an explicit applicability rationale.
Meta Mode · Proposed Grading Anchors
1–2/10
Purpose and scope are emerging.
3–4/10
Fields, contracts and examples exist; operation is limited.
5–6/10
Linked records support use, with manual checks and incomplete validation.
7–8/10
Repeatable execution, checks and dependency handling are demonstrated.
9–10/10
Independent evaluation and sustained reproducibility are demonstrated.
These anchors define a proposed rubric. Applying it consistently and validating its usefulness remain open work.
Mettā Mode · Proposed Grading Anchors
1–2/10
Care and agency requirements are being identified.
3–4/10
Explicit care, consent and attribution criteria are documented.
5–6/10
Representative outputs show those criteria being applied with gaps recorded.
7–8/10
Diverse user review and observed accessibility, agency and care outcomes support the assessment.
9–10/10
Independent, sustained evaluation supports those outcomes across contexts, with correction and redress.
These anchors define a proposed rubric. Applying it consistently and validating its usefulness remain open work.
AkashicSTATE Applied To AkashicSTATE
A State Record That Includes Its Own Limits
Version
v0.1.0-draft.1
Classification
Metadata & Maturity Module
Edition
Iron · Defined Components
Overall Maturity
3/10 · Provisional Self-Assessment
Meta / Mettā
Not Assessed / Not Assessed
Implemented Here
Definition, record template, classification and exportable reference data
Still Required
Executing record service, validation, immutable history and independent rubric testing
Pass & Deployment Dates
Unassigned in the reusable record until supported by a receipt
NEXUS is the orchestra as a whole. Components are its instruments, workflows are its scores and a run is a performance. People remain the conductors, setting purpose and approving the outcome.
Mettā EXPAND² and explicitly defined procedures used within a lens
Metadata Module
AkashicSTATE
Template
A reusable webpage brief, evidence appraisal or publication structure
Artefact
An output: an appraisal, visual, PDF, page or post
Run
A particular execution using recorded inputs and component versions
Below Operational, And Beyond
Source & Data. Original wording, papers, images and source locators.
Action. A discrete step, such as capturing a citation.
Task. Steps serving one purpose, such as appraising a study.
Operation. A coordinated unit of work, such as an article evidence review.
Workflow. An ordered route through stages with input/output contracts.
Orchestration. Coordination across workflows, components and outputs.
Meta. Description and assessment of the system and its work.
Mettā applies across every layer. These layers describe the scale of work; they are separate from the seven material editions and the maturity rubric.
Portable Today · Executable In A Future Release
A Distributable Foundation
The reference package includes this webpage, the workflow map, sigils, component guidance, preserved source locators, rating definitions, an AkashicSTATE template and three proposed test cases.
This is a static reference package and deployment blueprint. It does not install a ChatGPT plugin, execute model calls, connect private accounts or publish on someone’s behalf. Source destination pages have not been freshly re-audited for this export.
Use It Now
Unzip the package and open editions.html.
Browse index.html for the full workflow and reusable guidance.
Choose a test case, supply the inputs and use your chosen model or tools.
Copy the blank state record, preserve original wording and record actual checks.
Keep proposed actions, applied changes and observed results separate.
Public reference pages work locally. The hosted ZIP download and external sources require network access; private workspaces may also require sign-in.
The Future Executable Package
Registry: component versions, prompt sources, schemas, dependencies and content hashes.
Runner: selected routes, model adapters, structured outputs, retries and resumable runs.
Validation: claim/source checks, asset freshness and actual browser or deployment evidence.
Controlled Delivery: account access, HOLD, approvals where required and publication receipts.
Plugin Interface: callable tools plus reusable instructions and optional UI.
These are planned capabilities with no committed deployment dates.
Could It Become A ChatGPT Plugin?
Yes. OpenAI’s current plugin architecture combines skills, MCP tools and optional interfaces. A future NEXUS plugin could expose the runner and state records through those tools. A ZIP of reference files alone does not implement that integration.
Documentation checked 19 November 2026. Availability and publishing requirements should be rechecked before implementation.
Proposed Acceptance Tests
Three Real-World Uses To Test
These cases have not been run. They test workflow usefulness and reliability. Commercial demand, willingness to pay and profitability require separate evidence.
NEXUS-TC01 · Not Yet Run
Research-To-Publication Setup
Input: One agreed research question, a supplied source set and a target webpage brief.
Outputs, Pass Criteria & Comparison
Expected Outputs
Source-linked appraisal, webpage draft, PDF draft, visual brief and Reddit/X drafts.
Pass Criteria
Every checkable claim has a locator or is flagged unresolved; versions and credits match; drafts and actual delivery are distinguished; missing-source input triggers HOLD.
Baseline Comparison
Use the same input and source access for a manual baseline. Record review time, unresolved claims, correction effort and independent reviewer disagreements.
Actual Result: Not Recorded Run Date: Unassigned
NEXUS-TC02 · Not Yet Run
Publication Quality Review
Input: One existing article, its source list, image credits, embeds and intended audience.
Outputs, Pass Criteria & Comparison
Expected Outputs
Before-and-after report, discrepancy register, proposed fixes and separately recorded browser checks.
Pass Criteria
A prepared fixture exposes a missing source, missing credit, stale asset, broken embed and duplicate placement; every planted issue is flagged or explicitly recorded as missed. No browser pass is inferred from text review.
Baseline Comparison
Compare the baseline and NEXUS review against a pre-agreed issue key. Count issues caught, missed, introduced and falsely flagged, plus time and correction effort.
Actual Result: Not Recorded Run Date: Unassigned
NEXUS-TC03 · Not Yet Run
Private NEXUS Workspace
Input: Two test accounts, one restricted sample project, agreed roles and a versioned source change.
Outputs, Pass Criteria & Comparison
Expected Outputs
Configured workspace specification, role matrix, templates, onboarding and rerun history.
Pass Criteria
Unauthorised access is denied; HOLD blocks delivery; retries do not duplicate publication; a source change marks dependent outputs stale; previous records remain available; unknown dates stay blank.
Baseline Comparison
Run the same acceptance checklist twice and record support interventions. Access and deployment tests require an implemented sandbox; documentation alone cannot pass them.
Actual Result: Not Recorded Run Date: Unassigned
Assessment Basis & Source Trail
What These Labels Establish
The ratings are provisional design judgements recorded in the 19 November 2026 discussion. Existing component definitions and versions are preserved from the published NEXUS source snapshot. Unknown editions, missing checks and future dates remain unresolved.
The Bronze sigil is original AI-generated artwork. The orchestra diagram is a native vector architecture map. Neither claims traditional attribution. Names, quoted source material and third-party content retain their existing rights; distribution is not a blanket third-party reuse licence.
18 classified records · Existing provisional scores retained · FOOTPRINT and TIER unassessed · Meta and Mettā unassessed
Public Release & Restricted Development
The Source Atlas implementation and operational source records remain in the owner-only sandbox. Public citations and credits remain visible. Advanced source adapters, connected execution, managed reviews, telemetry and team operations require separate development and access controls.
After every AkashicNEXUS or AkashicSTATE run, including reruns, finish with a Post-Run Human Review handoff. Invite Jatinder Sahota to review as Creator / Owner-Reviewer. This is a separate human review; creator involvement means it is not independent external validation. Offer an answer-free packet where suitable and record whether answers were actually hidden, seen or unknown. Record the run ID and edition, scope, reviewer role, method, date, judgements, disagreements and reasons, corrections and rechecks. Keep Owner Review and Independent External Validation as separate statuses; use Pending until actual evidence supports another status. Preserve dated before-and-after records in AkashicLEDGER or the current versioned record. Owner review may support a justified AkashicSTATE reassessment but never raises ratings automatically. Respect the authorised audience, private source metadata and existing HOLD boundaries. This adds a required handoff, not a new publication permission gate.
Required Closing Record: Owner Review: Pending / Completed / Declined / Deferred. Independent External Validation: Pending / Completed / Not Applicable (with reason). Include the actual reviewer role and evidence; never infer completion from silence.