MurMur · Vision

Put capital in your context.

A personal capital office for everyone, with useful collective understanding.

Product design in development. Research and simulation; no investment execution.

MurMur · your own circumstances and choices. Concept illustration, not a live portfolio or investment performance.

Mission

Turn holdings, evidence and objectives into maintained understanding and reviewable capital choices.

One of Skyzai’s seven product strategies, designed to contribute to outcomes for Me (individuals), We (organisations) and They (machine intelligence).

Rosetta agent row · Theology

What is treated as ultimate, and with what warrant?

MurMur relates research and capital choices to a person’s enduring goals while comparing independent perspectives without erasing their differences.

Objectives

Start with your goals, holdings, obligations and constraints.
Research opportunities in the context of your own mandate.
Keep decisions, corrections and research contributions under your control.

Strategy

Connect independent offices through permissioned research, disclosed references and controlled shadow workflows.

Your personal office

Bring your position, protected reserves and goals into one maintained view.

Opportunity research

Inspect a venture or strategy before deciding whether it fits.

Independent research network

Choose whether to compare disclosed research from other offices.

KPIs

Proposed measures of progress

Comprehension, constraint preservation, aggregate accuracy, correction fidelity and transparent costs.

These measures describe how to evaluate this product’s contribution. Each needs a defined scope, observation period and evidence.

Targets: Not set. Thresholds remain to be agreed for the proposed measures.
Performance observations: No measured results are recorded here for these proposed measures.

Each proposed measure has a declared review population and evidence owner. Targets are unset and observations are Not recorded. Website test passes remain separate from customer performance.

Decision comprehension

Eligible population: Consenting participants completing the personal capital-choice study; incomplete sessions are reported separately.
Calculation: Participants correctly explaining reserves, uncertainty, full cost and execution authority / all completed study sessions.
Period: Each predeclared usability study.
Evidence source: Consented task responses and an independent reviewer’s rubric.
Accountable role: Product research lead; appointment unassigned.
Limits and exclusions: A fictional case and small sample do not establish financial literacy, suitability or investment outcomes.
Target / observation: Unset / Not recorded

Mandate fidelity

Eligible population: All reviewed capital-affecting research choices in the declared review sample, including held, declined and unresolved choices.
Calculation: Choices whose disposition preserves every current purpose, horizon, reserve, obligation and cost constraint / all reviewed choices. Report held, declined, unresolved and stale versions separately.
Period: Each release candidate for fixtures; monthly for any separately qualified service.
Evidence source: Versioned mandates, sources, decision trails and independent case review; fixture checks are labelled separately.
Accountable role: MurMur product reviewer; appointment unassigned.
Limits and exclusions: Passing known constraints does not prove affordability, solvency or a safe investment. We concerns permitted owner research; They uses a principal’s limits.
Target / observation: Unset / Not recorded

Collective reference fidelity

Eligible population: All fields and source links in a declared, separately permitted reference set; excluded or withdrawn contributions counted separately.
Calculation: Reference fields and lineage links matching the permitted, current source set / all checked fields and links. Report duplicated origins and unknown independence.
Period: Each reference revision and correction review.
Evidence source: Permission records, contribution versions and independently recomputed reference fixtures.
Accountable role: Research-data reviewer; appointment unassigned.
Limits and exclusions: This case has one synthetic contribution and calculates no aggregate. Derivation fidelity does not measure investment truth; privacy and inference leakage need separate qualification.
Target / observation: Unset / Not recorded

Correction and withdrawal fidelity

Eligible population: Every affected current choice, contribution, reference and export in the declared correction/withdrawal test matrix.
Calculation: Affected views correctly marked stale, excluded or explicitly refreshed / all affected views. Preserve earlier records and disclose unconfirmed downstream handling.
Period: Every correction/withdrawal suite and any later incident review.
Evidence source: Versioned before/after records, permission history, exports and independent review.
Accountable role: Privacy and evidence reviewer; appointment unassigned.
Limits and exclusions: Local behaviour cannot prove downstream deletion, lawful processing or revocation at a live recipient. Withdrawal must preserve personal-office use.
Target / observation: Unset / Not recorded

Cost transparency

Eligible population: All eligible proposal/choice views in the declared case and build-profile matrix, including incomplete prices.
Calculation: Views showing currency, full known cost, term, ceiling, uncertainty and unsigned status—or explicitly blocking a missing price—/ all eligible views.
Period: Each website candidate and later material terms change.
Evidence source: Rendered narrow/wide review and versioned offer/mandate records.
Accountable role: Product acceptance reviewer; appointment unassigned.
Limits and exclusions: Synthetic USD costs are illustrative. No live quote, billing, savings or investment performance is measured.
Target / observation: Unset / Not recorded

Launch evidence

Dated measurements of product launches, live integrations and published customer outcomes have not been recorded here. Missing evidence is not a measured zero.

Released services: Not recorded
Live product integrations: Not recorded
Published customer outcomes: Not recorded

Historical declaration · 18 October 2026. Owner direction, 18 October 2026. An owner-specified zero-release baseline was recorded for planning. It is retained as source history and is not a measured product result.

Agents

The Rosetta agent row connects this product’s field of inquiry to occupations, skills and responsibilities. The occupations below describe specialist work; different people, models and tools may carry these capabilities within an explicit assignment.

Rosetta agent row. Product L7 · Theology

Responsibility. Personal AI family offices and collective orientation

Reasoning approach. Reflective critique: question the assumptions behind a long-term goal, examine rival interpretations and revisit the horizon as circumstances change. Intuition can suggest a question; evidence must still support a claim.

Mandate researcher

Skills and responsibilities. Clarifies horizon, liquidity needs, constraints and the questions a decision must answer.

Diligence analyst

Skills and responsibilities. Compares sources, assumptions and alternatives, including reasons to wait or decline.

Research synthesist

Skills and responsibilities. Explains collective references and their origins while preserving personal limits. Participation does not pool assets or authorise trades.

Participants and supporting research, engineering and review capabilities; delivery appointments remain open.

What each specialist receives and returns

Mandate researcher contract

Inputs: Permitted goals, dated position, obligations, horizon and protected reserve.
Skills: Clarify purpose; check scope, cash limits and changed circumstances; surface missing records.
Output: A versioned mandate, constraint assessment and unresolved questions.
Limits: No keys, asset custody, treasury appointment, suitability certification or authority to spend.

Diligence analyst contract

Inputs: The current mandate and separately permitted, attributable offer/evidence records.
Skills: Compare alternatives, costs and terms; trace source ancestry; examine uncertainty and reasons to wait or decline.
Output: Attributed research and an unsigned choice with assumptions, costs and review triggers.
Limits: Repeated model output is not corroboration. No trade, investment recommendation service or performance claim is established.

Research synthesist contract

Inputs: Only purpose-bound, separately permitted contribution fields and correction/withdrawal records.
Skills: Check comparability, current permission, common sources, revisions and inference-leakage risks.
Output: A disclosed reference with provenance, exclusions, uncertainty and an auditable correction trail.
Limits: No automatic private-context access, training consent, pooled money, independent-source claim or permission to execute.

These are proposed capabilities, with appointments and live acceptance open. Product L7/Theology is a reflective correspondence, not a network layer, staffed seat or grant of permission.

These are roles and skills in development, with a defined context and an output to inspect. Their performance needs evidence and evaluation; a role description does not give an agent permission to act.

A decision starts with your circumstances.

Nora is considering research into workshop equipment. Does the cost fit the circumstances she has recorded?

Fictional local example. All figures are synthetic USD amounts; no account is connected. This is separate from Nora’s Me story and the workshop’s procurement case.

Your purpose and limits

Investigate equipment for Nora’s independent workshop within 12 months.

Research only. No investment, order, custody or transaction authority.

Recorded cash: USD 10,000
Protected reserve: USD 5,000
Existing obligations: USD 2,000
Uncommitted cash after reserve: USD 3,000

Position source: fictional self-entered cash/obligation record, revision 1. The reserve is a limit for this example, not a recommended reserve amount. Arithmetic alone cannot establish affordability or suitability.

Compare a research option

Full illustrative cost: USD 600. Commitment: 1 month. Research ceiling: USD 800.

Offer source: Fictional maker report quote, revision 1

Uncertainty: Maker claims are uncorroborated. Financials, equipment performance and future resale value are unverified. A collective preference would not resolve these gaps.

Review: Not reviewed

Uncommitted cash after this cost: USD 2,400. Review the current version before recording a choice.

No current decision. Any earlier choice remains in the history.

“Within the illustrative mandate” checks only the displayed limits. It creates no recommendation, signed instruction, trade, payment or provider appointment. A real purchase needs its own current terms and applicable authority.

Research participation is a separate choice.

Your personal office works with research permission off. Personal use, an unsigned draft and a group membership never opt you in.

Purpose: Compare research methods in this fictional workshop case. Recipients: A simulated comparison view on this page only.

Permitted fields: case revision, offer label, research disposition. No name, cash, holdings, obligations, keys, wallet address or contact graph.

Retention: Page memory until reload or reset; download is a separate local copy. Withdrawal removes the current simulated contribution without removing the personal decision. This page sends nothing; a future service cannot promise to recall independently retained copies. Omitting names does not prove anonymity or prevent inference leakage.

Permission: off. No contribution in the current comparison.

Inspect the versioned case and decision trail

The export includes the fictional personal position, uncertainties, earlier decisions and separate permission history. It is an unsigned local record.

{
  "type": "fictional-murmur-case",
  "synthetic": true,
  "unsigned": true,
  "maySign": false,
  "mayAuthorize": false,
  "mandate": {
    "goal": "Investigate equipment for Nora’s independent workshop within 12 months.",
    "currency": "USD",
    "cash": 10000,
    "reserve": 5000,
    "horizonMonths": 12,
    "researchCeiling": 800,
    "permission": "Research only. No investment, order, custody or transaction authority."
  },
  "revision": 1,
  "offer": {
    "name": "A bounded maker report",
    "cost": 600,
    "months": 1,
    "source": "Fictional maker report quote, revision 1"
  },
  "assessment": {
    "obligations": 2000,
    "available": 3000,
    "afterCost": 2400,
    "conflicts": [],
    "gaps": [],
    "status": "Not reviewed"
  },
  "uncertainty": "Maker claims are uncorroborated. No verified financials, investment return, suitability or solvency assessment.",
  "decision": null,
  "history": [],
  "research": {
    "scope": {
      "purpose": "Compare research methods in this fictional workshop case.",
      "recipients": "A simulated comparison view on this page only.",
      "retention": "Page memory until reload or reset; download is a separate local copy.",
      "fields": [
        "case revision",
        "offer label",
        "research disposition"
      ],
      "exclusions": "No name, cash, holdings, obligations, keys, wallet address or contact graph."
    },
    "permission": "off",
    "current": false,
    "contribution": null
  }
}

Download the fictional case

Reload also resets this page. Reset cannot erase a file you downloaded. No personal input, conversation, telemetry, consent or real research is stored or sent by this case.

MurMur is optional. Organisational research does not appoint a corporate treasurer; machine work does not create an autonomous treasury. Agentz can prepare a scoped proposal.

Start with your own circumstances.

MurMur is designed as a network of personal AI family offices. Participants run their own offices. MurMur’s proposed personal office brings goals, assets, liabilities, commitments and permitted evidence into a maintained understanding. A useful capital choice starts with your circumstances.

You can use the personal experience without contributing to collective research. Your goals, time horizon, liquidity needs and limits remain explicit. Agentz can help prepare an unsigned proposal; your office keeps the reasons to continue, wait or decline.

How proposals work

Let independent work inform a wider view.

MurMur’s collective intelligence is designed to connect separately permitted research from independent offices and the other products.

Concept diagram: independent offices and comparison groups, not pooled funds or live portfolios.

apu.bot organises source material against a purpose.
Circle contributes sourced observations and corrections.
Reality Futures contributes scenarios, forecasts and uncertainty.
Agentz prepares reviewable plans and proposals; Cortex preserves accepted memory and Nexus scopes permitted context within Agentz.
Menexus contributes preparation, specified protection and inspection of commitments and outcomes.
LIQUID defines separately governed market and transaction capabilities.

Compare like circumstances: goals, risk, horizon, mandate and other constraints. Research views must disclose their origins, shared sources and revisions. A group’s preferred allocation is a reference to examine; it does not replace your own limits or judgment.

Using an app does not contribute your information automatically. Research needs a separate, purpose-bound choice, minimal data and defined correction and withdrawal handling. Agreement among contributors is not evidence of accuracy or investment performance.

Research, decisions and execution stay distinct.

The full ambition combines useful personal offices, opportunity research and disclosed collective references. The bounded website journey lets you inspect a fictional position and illustrative mandate, compare a maker case, record a reason to wait or decline, and export the result.

MurMur never holds assets, signs orders or executes transactions. Participants retain their keys and mandates. Research participation does not pool your money or authorise trades. Any execution belongs to a separately chosen provider acting under a valid mandate.

The current work is product design and bounded proofs. Live app integration, independent adoption, customer outcomes and commercial terms remain open. This page offers no account opening, investment subscription or order execution.

Research and progress · The proposal boundary

Back to the product family

Capabilities and current proof.

As reviewed on 2026-10-17, these are selected capabilities traced to their product owner. The proof level describes a plan, source observation or bounded test; it does not establish a released service.

Personal AI family office · Proposed
The owner assigns personal resource and mandate work to MurMur; CP-14 is a historical shadow-office proposal.
Current limit: No connected family-office service or delegated operating mandate is evidenced.
Separately permitted collective research · Proposed
Collective comparison/aggregation is part of MurMur direction only with separate permission.
Current limit: Private case data does not automatically transfer to a collective and no released aggregate is evidenced.

Inspect the full capability record and source references

Product plan, responsibilities and sources

Direction and delivery

Skyzai’s vision is an emergent Marketsphere. Its mission is to develop SPECTRE as an open operating system for it. Me, We and They describe the participant objectives, and each of the seven products is a strategy with its own KPIs and Agents. This nested product plan explains its direction, objectives, delivery work, measures and responsibilities; customer outcomes remain to be demonstrated.

Vision: A personal capital office for everyone, with useful collective understanding.
Mission: Turn holdings, evidence and objectives into maintained understanding and reviewable capital choices.
Objectives: Understand personal constraints, investigate opportunities, compare references and preserve decisions and corrections.
Strategies: Connect independent offices through permissioned research, disclosed references and controlled shadow workflows.
Measures & triggers: Comprehension, constraint preservation, aggregate accuracy, correction fidelity and transparent costs.
Agents & skills: Participants and supporting research, engineering and review capabilities; delivery appointments remain open.

Plan basis and open decisions

Current owner direction defines product responsibilities. Retained proposals inform delivery plans and engineering work. Signed charters, current operating mandates, staffing and measured baselines retain their separate standing.

The full personal-office programme remains the scope. The first milestone is a bounded synthetic office journey; later collective references and shadow workflows are separate proofs, not a reduction of the product to a research feed.
Participants independently choose their offices, contributions and execution providers. Collective references must disclose origins and preserve personal constraints, correction and withdrawal handling.
Commercial direction remains a proposal. Live billing terms, legal and data responsibilities, independent adoption and customer outcomes remain open.

These working documents require access to the corresponding private GitHub repositories.

Reviewed product proposal and retained delivery plan · 2026-10-09
Corresponding Rosetta L7 row · 2026-10-09

Rosetta agent row.

Selected product correspondence. MurMur occupies L7: personal objectives, evidence and collective patterns remain answerable to individual circumstances. Emergentism supplies the framework across every level.

Product L7 · Theology: Personal AI family offices and collective orientation
What is treated as ultimate, and with what warrant? MurMur relates research and capital choices to a person’s enduring goals while comparing independent perspectives without erasing their differences.

How the Rosetta agent row is used

Skyzai presents seven products: APU, Circle, ReFu, Agentz, Menexus, LIQUID and MurMur. Agentz is the continuing centre, with Cortex memory and Nexus connections. APU is the interface. Menexus means ensurance, insurance and assurance. Product roles preserve the underlying Rosetta coordinates. Emergentism is the framework across the whole. The Spectre network is a component within the wider Marketsphere programme; Permanent Composability supports independent reuse across the products.

The Rosetta agent row relates this product’s inquiry, reasoning and specialist skills to Emergentism. Several models, people or tools may contribute to the work. The philosophical correspondence, product responsibility and agent dispatch contract remain distinct; this page neither establishes deployed performance nor grants a model permission to act. This product direction does not migrate runtime contracts or amend signed source history.