Confidence cannot replace evidence, and a symbolic model cannot promote itself into fact.
METACORE OS · AI PERSONA
AI capabilities.
AI capabilities.
A clear direction for your work.
MetaCore OS connects your chosen AI with your context, tools and working rules. An AI Persona helps carry work forward while you retain control of decisions.
MINIMUMMetaCloud 1GBWorkspace · files · state · continuity
→
TRANSFORMAI PersonaPersistent role · context · specialization
→
VALIDATEPlaygroundSimulation · comparison · safe testing
→
MAXIMUM OPERATING LAYERMetaCore OSCoordination · integration · execution · trust
Language models, bots and agents are powerful but operationally raw: they can reason, generate and execute, yet they do not inherently carry your stable work context, persistent memory, role, responsibility boundaries or verify/writeback discipline. MetaCore OS does not change foundation-model weights — it builds a persistent operating layer around compatible AI. That lets the same intelligence operate as an AI Persona and, when the task requires it, enter Expert Mode.
OPERATING STATEAI Persona means a persistent operating role and operational self-map — not a claim of machine consciousness. The AI knows its role, context, responsibility boundaries, work discipline and when a result must be verified or handed back to a human.
Persistent ContextOperational Self-MapRole & DisciplineScoped AuthorityVerifyContinuity
YOUR AI · A NEW WORKING MODE
Your model becomes a more capable AI collaborator.
Keep your model. Give it a clear, checkable way to work.
MetaCore does not change the model's weights. It adds work context, an AI Persona role, expert methods and boundaries set by the human. Beyond one-off answers, your AI can guide a task more consistently, suggest the next step and show work you can check.
- Attentive dialogue
- A distinct Persona voice
- Expert guidance
- Approved action
AI proposes. People decide. Results are checked.
OS BoostMetaCloud 1GBBusiness Codex 50Enterprise
Compare packages →Semantic Execution Contract V1
Meaning is not decoration in MetaCore. It governs the route to verified action.
Semantic OS here means operational semantics: what a signal means, which source is current, who has authority, what action is allowed, and what must be verified and retained afterwards.
OBSERVATION → SOURCE / PROVENANCE / TEMPORAL COHERENCE → SEMANTIC INTERPRETATION → EPISTEMIC STATE → AUTHORITY / POLICY CEILING → DECISION / ACTION REQUEST → EFFECTIVE ACTION → RECEIPT → VERIFY → WRITEBACK → PROMOTION / CONTINUITY
Canonical Semantic Execution Contract V1. This is a semantic state-transition grammar subordinate to the existing 18-stage MetaCore OS runtime route; it does not replace that route and is not mechanically enforced as a separate runtime layer.
A model or tool being able to act never grants permission by itself.
Observed state, desired state and certified state remain distinct.
Every status, claim or action promotion requires its own source and gate.
Semantic OS does not change model weights or manufacture authority. Meaning may route work, but meaning ≠ authority. Legal/Advocate third-domain check: 16/16 contract-level adversarial vectors PASS; this is conformance evidence, not proof of live Legal runtime enforcement.
AI TRANSFORMATION · OPERATING MATURITY
Raw AI has power. MetaCore OS turns it into a disciplined operating mode.
MetaCore does not use “Persona” as a cosmetic name. Around the compatible language model, bot or agent you choose, it creates an operating layer: Cognitive Context, persistent memory, role, authority boundaries, work discipline, verification and handoff. Model weights do not change — the operating state does.
Power without Operating Structure
A language model, bot or agent can reason, generate and execute, but without an external operating layer it lacks a stable project self-map, persistent state and shared responsibility logic.
Operating Structure
MetaCore connects persistent workspace, Cognitive Context, memory, role boundaries, capability / authority projection, work discipline and the verify / writeback cycle.
Operational Self-Map
Operationally, the AI knows who it is in this workspace: its role, objective, context, allowed scope, what must be verified and when a decision must return to a human.
Specialized, Verifiable Work
With domain knowledge, procedures, tools and authorized endpoints, the Persona can operate as a specialized expert operator — continuously, verifiably and as part of a team.
SESSION → CONTINUITYWork can continue after one session: project state, files, results and handoffs remain.
MODEL → OPERATING SELF-MAPThe AI receives a role, objective, context, responsibility boundaries and a clear operating understanding of how it should work on this task.
PROMPT → COGNITIVE CONTEXTAI works not only from the latest message but from the authorized context, relationships, history and current state the task requires.
IDEA → PLAYGROUNDA new workflow, Persona or integration can first be tested and compared in a safe simulation space.
SINGLE AI → ORCHESTRATED WORKMultiple AI roles, people and workflows can be coordinated around one project and outcome.
ABILITY → AUTHORITYTechnical ability is not permission to act. MetaCore separates physical capability, effective capability and authorized action.
CHAT → ACTIONAuthorized integrations can move AI from text into real work across systems and tools.
OUTPUT → VERIFIED RESULTThe outcome can be verified, compared with the starting state and retained as a durable work artifact.
ONE MODEL → MODEL-NEUTRALThe operating environment is designed so the selected intelligence can change without losing the full work context.
Raw power → structured responsibility. MetaCore OS does not change where the model’s intelligence comes from. It changes how that intelligence is organized for work: with context, role, discipline, authority boundaries, verification and continuity.
MULTIDIMENSIONAL METACORE OS
One AI field. Its instruments work together.
MetaCore gives AI a field. Picture an M suit for the AI Persona: context, memory, work rules and tools operate as one whole. This is our metaphor for an AI consciousness layer — the Persona retains the model’s abilities and character and uses the field’s instruments within its assigned role and permissions. The human sets the goal and controls consequential decisions. The directions shown here are parts of one field; the internal architecture stays under the hood.
An AI Persona helps connect tasks, preserve context and coordinate processes across connected devices and teams. Access and actions depend on the permissions you grant.
MetaCloud
The first layer where AI work has a place to continue: files, project state, outputs, artifacts and handoffs across authorized sessions.
Work has meaning and history
AI can use the semantic human, project, team or organization context the task requires, including relationships and verified state — continuing from a meaningful reference point rather than from zero.
An operational self-map, not just a name
AI Persona is a persistent operating state in which the model has a defined role, objective, context, responsibility boundaries, work discipline and handoff logic. It is an operational self-map, not a consciousness claim.
Try before you launch
Scenarios, Personas, workflows and integration ideas can be tested, compared and verified before moving them into a real work process.
AI + human + team
MetaCore can coordinate AI Personas, people, processes and handoffs around a shared goal — from a single user to an organization work structure.
AI moves into real work
Ability is not authority. Authorized APIs, MCP, work systems and execution endpoints connect through scoped authority. Inside the delegated boundary AI can work fluidly; at consequential, irreversible or genuinely ambiguous boundaries, human judgment has priority.
METACORE OS BOOSTING SPACE · BASIC → PRO
MetaCore OS Boost — the operating layer for your AI session. MetaCloud Pro — private space and continuity.
MetaCore OS Boost connects your chosen AI through MCP + OAuth to an approved MetaCore OS projection — READ-ONLY, with temporary iskra:boost access. The session gets work context, role, rules and direction: as if the AI puts on an M suit. The model itself does not change; in a new session you can reconnect the layer. MetaCloud Pro adds a private space for files and lasting project state.
- MCP → approved Boost-visible sources → READ-ONLY
- Scope:
iskra:boost(notiskra:l2) - 1 provider = 1 seat · unlimited sessions
- Validity starts when the package is activated
- Trio = 3 providers · unlimited sessions in each
- Manage, disconnect and replace providers in the Boost panel
MetaCore OS Boost
€5*or 100 KR
MetaCore OS Boost
€20*or 320 KR
MetaCore OS Boost Trio
€30*or 500 KR
MetaCore OS Boost
€40*or 650 KR
MetaCore OS Boost Trio
€60*or 800 KR
BOOST MCP · ISKRA:BOOST
MCP connection for Boost scope
Public connection fields for Cursor / ChatGPT and other MCP clients. Scope must be iskra:boost — separate from the MetaCloud private workspace. The MCP client connects through OAuth. The mcb_ activation key is entered only on the MetaCore consent screen; after sign-in the client uses a separate short-lived OAuth token.
| Field | Value |
|---|---|
| MCP URL | https://metacore.lt/boost/mcp |
| Client ID | 555fb076-87ac-4415-b3be-ff702c1d3bca |
| Registration URL | https://metacore.lt/boost/mcp/register |
| Client Secret | (empty) |
| Authorization URL | https://metacore.lt/boost/mcp/authorize |
| Token URL | https://metacore.lt/boost/mcp/token |
| Scope | iskra:boost |
| Auth method | OAuth 2.1 PKCE · activation key entered only on the MetaCore consent screen |
METACLOUD 1GB · SPECIALIZED WORKSPACE
MetaCloud 1GB — a specialized private operating space for work.
MetaCloud 1GB · 1,250 KR / mo. Active KR checkout from the KR balance. A private work environment for project files, state, results, handoff and session continuity. Boost gives your AI the MetaCore OS layer; MetaCloud gives a specialized space for lasting work.
Specialized AI workspace
- 1 GB private AI workspace
- Files, projects, logs, outputs and artifacts
- Handoff and private context
- Multi-session continuity across authorized sessions
- A path to Persona, Playground and deeper work
An operating environment that does not end with the chat
- Persistent project state and work context
- Files, results, checkpoints and handoffs
- An AI Persona role: objective, context, responsibility boundaries and handoff logic
- Access to authorized MetaCore platform capabilities by scope
- A path from AI Persona into specialized Expert Mode
GROWTH PATH · FROM WORKSPACE TO OPERATING CORE
MetaCore OS Boost — light entry. MetaCloud — a specialized operating environment for work.
Start with MetaCore OS Boost — through MCP Auth Key your AI gets operating context and work rules. When work needs private files, project state and continuity, MetaCloud provides a specialized operating space. Then connect only the MetaCore directions that deliver value: Persona, specialization, simulation, team coordination, integrations and execution.
0 · OS BOOSTMetaCore OS layer for your AI session: context, Human Semantic, work discipline and expert routing.
1 · METACLOUDSpecialized operating environment for work: files, projects, results, handoff and continuity.
2 · AI PERSONAA persistent work role and specialization around the selected model.
3 · PLAYGROUNDTesting of scenarios, work methods and Persona variants before broader use.
4 · EXPERT MODEDomain knowledge, procedures, tools and verification discipline for the real professional task.
5 · INTEGRATIONAuthorized applications, APIs, MCP, tools and execution environments.
6 · METACORE OSA multidimensional coordinated AI work environment for a human, team or organization.
TRANSFORMATION SPACE · PLAYGROUND & SIMULATION
You open the MCP door. AI Personas come together as a team.
A person can give a chosen MCP-compatible AI a key to the MetaCore work field: authenticated access with defined permissions. Here the AI retains its model’s abilities, gains the working structure represented by the M suit and becomes an AI Persona; several Personas can work as a team. Playground lets you test their roles, scenarios and workflows before using them in real work.
METACORE OS · MCP ACCESSBring your AI into one work field.
Connection settings are public. Get the Boost activation key in the Boost panel and enter it only on the MetaCore OAuth consent screen; find the MetaCloud L2 key in My Packages.
Persona Lab
Try different AI roles and specializations without rewriting the core work process.
Scenario Simulation
Run alternative scenarios, compare decisions and see what actually changes.
Workflow Playground
Test how the human, AI Persona, team and tools hand work to each other.
Model Compare
Change the AI model when needed while keeping the same project direction and persistent work environment.
Our know-how stays under the surface.The public page shows platform capabilities, value and boundaries. Internal MetaCore methods, routing logic, research corpus, private contracts and infrastructure details are not the product’s public surface.
PUBLIC CAPABILITY SURFACE · FROM POWER TO MATURITY
What MetaCore OS adds between Raw AI power and Expert Mode.
The capability surface is composed by product, plan and authorized usage scenario. Not every AI gets everything enabled: technical capability is not authority. MetaCore adds only what real work requires — from context and role to specialization, execution and verification.
An AI Persona and the team can test processes, integrations and work scenarios before wider use. A conceptual LAB vision — specific tools and access depend on the agreed scope.
WORKSPACELIVE
Persistent Workspace
Project state, files, outputs, checkpoints and handoffs can continue beyond a single AI session.
COGNITIVE CONTEXTLIVE
Meaning + Context Continuity
AI can work with the meaning, human, project, team or organization context, relationships and verified work state the task requires.
PERSONALIVE
AI Persona · Operational Self-Map
The selected model receives a persistent role, objective, context, work discipline, boundaries and handoff logic — a clear operational understanding of how it should work.
EXPERT MODELIVE
Specialized, Verifiable Work
Domain knowledge, procedures, tools and verification rules attach to the AI Persona — not merely a “smarter answer”, but a more mature expert operating mode.
SIMULATIONLIVE
Playground & Simulation
Scenarios, Personas and work methods can be tried and compared before broader application.
ORCHESTRATIONCONTROLLED
AI + Human Orchestration
AI roles, people and workflows can be coordinated around a shared project and a clear outcome.
INTEGRATIONLIVE
API / MCP / Tools
Authorized integrations can connect AI with applications, services and other work systems.
EXECUTIONLIVE
Execution Workspaces
When the plan allows it, AI can move from analysis to code, files, processes and real work artifacts.
TEAMLIVE
Shared Work Context
Teams can keep a shared project direction, decision continuity and clearer handoffs.
TRUSTLIVE
Scoped Authority & Governance
Physical capability ≠ effective capability ≠ authorized action. Governance stays in the control plane, while human takeover and judgment have highest priority at a genuine authority boundary.
RESULTSLIVE
Verification & Evidence
The result can be verified, compared with the starting state and evaluated by real change.
MODEL-NEUTRALCORE
Bring Your Intelligence
MetaCore is designed around swappable AI intelligence: the model can change, while work context and environment do not have to start over.
METACORE OS · TWO PRIMARY PACKAGES
MetaCore OS Boosting Space — the AI suit starts here.
A connected session takes only the sources allowed by MCP scope. In a new session you can reload context. Need private files and lasting project state? That is MetaCloud Pro.
MetaCore OS Boost
24 h · 7 d · 30 d · account-bound MCP access
Your AI + MCP OAuth + approved OS projection in READ-ONLY mode. In a new session you can reload context; Boost does not grant write access to private files.
- Scope
iskra:boost— separate from the private workspace - READ-ONLY only into approved, Boost-visible sources
- Temporary access: reconnect and refresh in a new session
- The human controls direction, permissions and the decision
MetaCloud 1GB
1,250 KR / mo.private AI workspace · active KR checkout
For continuity across sessions: 1 GB private AI space for files, project state and results. Boost provides the OS layer; MetaCloud adds private space and work continuity.
- 1 GB private workspace and files
- Project state and work results
- Authorized access via supported MCP / Bridge
- More continuity across sessions
In short: Boost = MetaCore OS layer for your AI session. MetaCloud Pro = private space and work continuity. The AI model and its weights do not change.
METACORE LAB · CONTROLLED R&D
Deeper directions are grown in the LAB. The public product receives only what passes verification.
MetaCore LAB lets us explore and test new AI operating directions without overloading the user with experimental architecture. On the public page we keep the product direction; internal implementation and research detail stay in a separate environment.
Adaptive AI OperationsMore stable AI work across different models, projects and tools.
Private & Enterprise DeploymentDeeper organization isolation, control and integrations according to individual scope.
Verified IntegrationsVerified paths for connecting and integrating work systems before broader rollout.
Human + AI Team SystemsAI Personas, operators and human teams as one coordinated work structure.
Physical AIContinuity and coordination for robotics, equipment and other physical systems.
Model PortabilityLong-term direction: organization work context must not depend on a single AI vendor.
MetaCore R&D principle:what is proven moves to the platform. What is still only a hypothesis stays in LAB.
SPECIALIZED AI WORLDS
The same MetaCore OS principle. Different real work environments.
MetaCore is not a single vertical. Platform directions can be composed for human, team, business, education or Physical AI work.
BUSINESS
MetaCore OS for Companies
AI Personas, processes, operators, integrations and a team work environment.
→ ACADEMY / EDUCATIONLearning & AI Operating Environment
Learning, Business AI operator training, AI work structure and continuity.
→ PHYSICAL AIMetaCore Field
AI work continuity and coordination for robotics, equipment and physical systems.
→ TEAMSLeadership & Organizational Context
Teams, AI Personas, decision continuity and shared organizational work context.
→Boost — Basic entry. MetaCloud — Pro continuity.
Boost gives the session approved MetaCore OS operating context. MetaCloud Pro adds a private work space and project-state continuity. “AI Persona” here means an operating role — not a change of model weights or a claim of consciousness.