Conversation
Speak or type your way from scattered thoughts to a clear question.
apu.bot · Vision
Your case, wherever you are.
Your ideas, questions and work stay within reach, wherever you are.
Explore apu.bot
Product design in development. What you can explore today.
Mission
Provide a workspace that captures permitted material, clarifies purpose, maintains a navigable outline and makes changes understandable and recoverable. Begin with one bounded personal workflow.
One of Skyzai’s seven product strategies, designed to contribute to outcomes for Me (individuals), We (organisations) and They (machine intelligence).
Rosetta agent row · Teleology
What end is being proposed?
APU helps you articulate a purpose, name constraints and recognise questions that remain open before work begins.
Objectives
- Make the purpose clear before working on an answer.
- Keep sources, suggestions and decisions understandable.
- Help you return, revise, undo and export your work.
Strategy
Clarify intent and constraints; compare alternatives; present acting account, permitted context, consequences and review. The website demonstrates one fixed brief, account/context changes and correction locally. Voice, durable outlines, undo, export and cross-device continuity remain delivery work with Agentz.
A living outline
Move between the overview and the detail of one continuing case.
Work across devices
Return through the interface that suits the moment, with Agentz carrying permitted context.
KPIs
Proposed measures of progress
Define brief comprehension, context/action fidelity, suggestion/decision distinction, find/resume effort and correction/history fidelity. Each definition names its eligible population, calculation, period, evidence source, accountable role and limits. Observations are Not recorded and targets Unset. Undo/export fidelity and two independent hosts for component reuse need separate proofs.
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.
Proposed measurement definitions. Observations are Not recorded and targets are Unset. No collection is connected here; a fictional case is not product performance evidence.Brief comprehension · definition and limits
Context and action fidelity · definition and limits
Suggestion and decision distinction · definition and limits
Find and resume effort · definition and limits
Correction and history fidelity · definition and limits
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 · . 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 L1 · Teleology
Responsibility. Purpose, intake and organisation
Reasoning approach. Dialectical inquiry: clarify the purpose through questions, compare conflicting requirements and leave unresolved contradictions visible.
Intake guide
Skills and responsibilities. Clarifies purpose, asks for missing context and organises the material you choose.
- Inputs
- A stated need, permitted notes and explicit constraints.
- Output
- A proposed brief with its supported decision and unanswered questions.
- Limits
- Cannot invent missing facts, promote suggestions into decisions or commission work.
Information organiser
Skills and responsibilities. Connects notes and sources without turning a suggestion into an accepted decision.
- Inputs
- Attributed material, acting account, available context and current case state.
- Output
- A navigable outline that separates sources, alternatives, suggestions and reviewed choices.
- Limits
- Cannot expand access, disclose notes or treat a local review as external authority.
Continuity guide
Skills and responsibilities. Helps you find the right case, inspect changes and choose what to retain.
- Inputs
- Permitted case references, source revisions and an explicit retention choice.
- Output
- A return path and revision summary, with unavailable context visible.
- Limits
- Agentz owns Cortex memory and Nexus connections. Cross-device resumption and recovery need separate implementation evidence.
People supply purposes and appropriate acceptance. Bounded software performs intake and organization. Engineering, integration and independent evaluation capabilities need actual appointments; the documentation does not fill them.
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.
See it in a piece of work.
Local demonstration · independent-analysis case
From a loose need to a brief you understand.
The workshop wants independent analysis before deciding how to develop its robotics project. Try a personal, organisation or machine-task view of that need. These are fictional example accounts; selecting one connects no account or mandate.
1 · Make the brief understandable
- Purpose
- “We need independent analysis.” Which decision it should support is still unresolved.
- Material visible to this example account
- The team-approved, redacted project notes · revision 1. Timing is not yet supplied.
- Cost and permission
- USD 200 proposed team ceiling. This is a fictional constraint, with no funds held. Access to notes gives no permission to disclose them, commission work or spend.
The example supplies fixed wording so you can inspect the states. It generates no AI answer.
2 · Compare, review and choose
- Visible action
- Clarify first · the workshop’s analysis brief
- Acting account and context
- Workshop · organisation · Current project material permitted · source revision 1
- Possible recipient
- A possible independent analysis provider. No recipient has been selected or contacted.
- Consequence
- List the missing decision, deadline and constraints. The need stays open. Keeping the choice changes only this page’s example.
Suggestion · review required · Workshop · organisation · Current project material permitted.
Changing the account, permission, source or alternative clears the current review. Earlier local choices retain their original scope in history. Returning to an earlier setting requires a new review.
3 · Choose a communication entry
After keeping a permitted choice to clarify or prepare, inspect a draft addressed to the possible recipient. Recipient selection, disclosure review, consent and actual delivery belong to a connected service.
A draft is an entry to communication, not a sent request. Closing discards it. Changing context also closes it; reset starts a fresh example.
Correction, interruption and local history
A corrected source needs fresh review. Interruption pauses the example; resuming adds no access. Neither creates cross-device continuity or a recovery policy.
No local choices recorded.
Everything here lives only in this page’s memory and resets on reload or leaving the page. It sends no request, stores no conversation, opens no service and moves no money. Agentz owns the proposed continuing Cloud Computer, Cortex memory and scoped Nexus connections.
One job, with clear boundaries.
- Starts with
- Your purpose, notes, links and conversations, with permission to use the material.
- Produces
- A navigable case and conversation, with sources, proposed changes and accepted decisions visibly distinct.
APU preserves the distinction between a possibility, a proposal and an accepted decision. Ordinary notes can be organised under your declared consent; external actions retain their own acceptance boundaries.
Capabilities and current proof.
As reviewed on , 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.
Voice, text and navigation UI · Source observed
Flutter voice-capture and search/navigation source exists in the company app.
Current limit: Source presence does not prove complete device journeys, speech quality or an independently released APU UI.
Continuing case and source presentation · Bounded local proof
The shared-case page and local preview present records, citations and review controls; bounded browser and widget checks are recorded.
Current limit: The case UI is a composed host and its local preview does not establish a production signed-in release.
Local attributed case lookup · Bounded local proof
The local host calls the pinned APU librarian find/cite path against selected case records.
Current limit: Deterministic local retrieval is narrower than APU search service or live model analysis.
Typed ordering and reference policy · Bounded local proof
A standalone pure Dart ordering package and two fixture hosts demonstrate explicit criteria and tie handling.
Current limit: Synthetic reuse is not production extraction, independent-team adoption or the full APU interface.
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
- People can turn scattered ideas and information into purposeful work while retaining their sources, uncertainty and control.
- Mission
- Provide a workspace that captures permitted material, clarifies purpose, maintains a navigable outline and makes changes understandable and recoverable. Begin with one bounded personal workflow.
- Objectives
- Make purpose explicit or visibly unresolved; preserve attribution and distinctions between possibilities, proposals and decisions; help people resume and revise work; support reliable correction, undo and export.
- Strategies
- Clarify intent and constraints; compare alternatives; present acting account, permitted context, consequences and review. The website demonstrates one fixed brief, account/context changes and correction locally. Voice, durable outlines, undo, export and cross-device continuity remain delivery work with Agentz.
- Measures & triggers
- Define brief comprehension, context/action fidelity, suggestion/decision distinction, find/resume effort and correction/history fidelity. Each definition names its eligible population, calculation, period, evidence source, accountable role and limits. Observations are Not recorded and targets Unset. Undo/export fidelity and two independent hosts for component reuse need separate proofs.
- Agents & skills
- People supply purposes and appropriate acceptance. Bounded software performs intake and organization. Engineering, integration and independent evaluation capabilities need actual appointments; the documentation does not fill them.
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.
- Voice, earbud and phone integrations need their own latency, accessibility and recovery proofs. A visible composer does not prove a live service.
- Retention follows your settings and source permissions; presenting memory does not make APU its storage owner.
These working documents require access to the corresponding private GitHub repositories.
- Adopted owner product direction · 2026-10-13
- Reviewed product proposal and retained delivery plan · 2026-10-09
- Corresponding Rosetta L1 row · 2026-10-09
Rosetta agent row.
Selected product correspondence. APU remains the L1 product home for purpose and intake. Its interface reaches the shared case; durable memory and redundant storage belong within Agentz.
- Product L1 · Teleology
- Purpose, intake and organisation
What end is being proposed? APU helps you articulate a purpose, name constraints and recognise questions that remain open before work begins.
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.
Explore the design.
Connected capabilities in this website home. These are designs with distinct service responsibilities.
These pages describe the intended product and its delivery work. They do not open a signed-in app or establish operating availability.