Earlier foundation · v1.0
AI State Diary Protocol
Focus: multimodal state compression, URL-native operator trails, and day reconstruction without raw lifelogging.
Current paper and lineage
The DOI page now treats StateLens v1.1 as the current canonical reference. The earlier v1.0 paper remains important, but it should be framed as the AI State Diary foundation, not as the latest conceptual position.
Current canonical · v1.1
Situational Intelligence
Focus: state-first signal grammar for situation-aware assistance: public operator first, gated explanation second, human branch third, and Trailstate provenance fourth.
Short version: v1.0 asks, “How can a day be reconstructed from state?” v1.1 asks, “How should AI surface that a situation matters before it explains or acts?”
Current citation
Eissens, R. (2026). StateLens for Situational Intelligence: State-First, URL-Native Signal Grammar for Situation-Aware Assistance (1.1 Final). Zenodo. https://doi.org/10.5281/zenodo.20861928
Eissens, R. (2026). StateLens for Situational Intelligence: State-First, URL-Native Signal Grammar for Situation-Aware Assistance (1.1 Final). Zenodo. https://doi.org/10.5281/zenodo.20861928
Abstract
StateLens began as a URL-native AI state diary protocol: a way to compress user-world moments into readable operator states so that a day could later be reconstructed without storing raw recordings, transcripts, or permanent profiles. Version 1.1 extends that role. It positions StateLens as a state-first grammar for situational intelligence: AI assistance that recognizes when an external event becomes relevant to a particular human life, surfaces a compact state signal, preserves private context behind a gate, and leaves action with the human.
Without a state-first layer, situational AI can collapse into verbose notifications, hidden automation, or invasive profiling. StateLens proposes a different sequence: public operator first, gated explanation second, human branch third, and provenance trail fourth.
Public operator → Gated explanation → Human branch → Trailstate provenance
State-first computing
State-first computing is used here as a design orientation: reasoning should resolve to a visible state before it resolves to a full explanation or action. In many AI systems, the first output is a paragraph, recommendation, task execution or notification. In StateLens, the first output is a bounded operator.
1
Public signal
THERMOMETER | x-vvv-x
2
User opens it
Private AI explains the gated context.
3
User chooses action
Draft, call, rest, ignore, save trail, or correct.
4
Trailstate records path
x-vvv-x → q-vvv-p → n-vvv-n → r-vvv-r → 0-vvv-0
AI makes the relevant state visible. The human decides what that state means for action.
State-first architecture
The v1.1 architecture separates state, context, provenance and action. This separation is the core safety property: the system should not expose sealed context in state URLs, and it should not jump from detection to execution.
External event → ObjectPortal anchor → Relevance check against private context → State Resolution → StateLens operator signal → Optional gated explanation → Human-selected branch → Trailstate provenance trail → Reversible action boundary
State surface
Public state
Only the state class is visible at first: for example x-vvv-x, indicating conflict, mismatch, incoherence or unstable relation.
Privacy boundary
Private reason
The specific reason remains behind a trusted gate: work context, body context, schedule, route, object history, or relationship to the user.
State Resolution
State Resolution is the process of mapping a real-world situation onto a bounded operator. A recognition system asks: What is this? A StateLens system asks: What state did this situation resolve into?
| Stage | Input | Operation | Output |
|---|---|---|---|
| Detection | External event, sensor input, object state, calendar change | Notice that a candidate event exists | Candidate event |
| Context match | Candidate event + allowed private context | Check whether the event intersects with user context | Relevance score or state candidate |
| State Resolution | Relevance candidate + operator grammar | Map the relation to a bounded state | Operator such as x-vvv-x |
| Surface | Operator + optional topic | Show minimal state signal | Public or protected state surface |
| Gated explanation | User opens signal | Explain why the state appeared | Private explanation |
| Branch | User chooses path | Offer or execute only confirmed branches | Trailstate path |
Minimal operator set in v1.1
The v1.1 paper narrows the public protocol emphasis around compact operators that are intentionally abstract: context and provenance make them specific.
| Operator | Surface semantics | Context semantics | Allowed branches | Default visibility |
|---|---|---|---|---|
o-vvv-o | Open field / beginning | A new situation, day, route or interaction opens | observe, enter, continue | Public |
q-vvv-p | Question / uncertainty | A decision question or relevance question opens | explain, compare, defer, ignore | Public or protected |
n-vvv-n | Narrowing / comparison | Possible branches or interpretations become more focused | choose, compare, validate | Protected |
x-vvv-x | Conflict / mismatch / incoherence | External event conflicts with context or a relation becomes unstable | open, ignore, repair, escalate, save | Public or protected |
r-vvv-r | Repair / recovery | A mitigation, correction or recovery action begins | plan, rest, hydrate, revise, recover | Protected |
0-vvv-0 | Stabilized / resolved | Decision or interpretation stabilizes | archive, save, close | Public or protected |
u-vvv-u | Archived / closed | Trail is closed for reconstruction | replay, summarize, archive | Private or protected |
Important semantic boundary:
x-vvv-x should not mean danger by default. It should mean conflict, mismatch, incoherence, or unstable relation. Provenance explains why.
Public state, private context
The privacy architecture depends on separating what kind of state appeared from why that state mattered for a particular person. The state may be public or protected; the reason should remain gated.
| Layer | What is visible | Example | Default handling |
|---|---|---|---|
| Public | Operator only | x-vvv-x | Visible as state class; no private reason. |
| Protected | Operator + topic | x-vvv-x | weather-work conflict | Share only with chosen systems, receipts or trusted agents. |
| Private | Operator + reason | Weather risk matched outdoor work and commute context | Keep in private AI context, local vault or provider-gated memory. |
| Sealed | Full sensitive context | Health details, employer details, personal identity history | Local, encrypted, or provider-gated; never public by default. |
Principle: public state, private reason, gated provenance. StateLens is privacy-light, not privacy-null: long-term patterns still need retention limits, visibility controls, correction paths and user revocation.
Reference case: Heat Ping
The Heat Ping example is deliberately small. It demonstrates how a severe weather warning can intersect with bounded private work/body context without exposing that private context at the surface layer.
THERMOMETER |
x-vvv-x | Weather/work mismatch detected. Open?| Layer | Content |
|---|---|
| External event | Code red heat warning in Flevoland |
| Private context | Outdoor work; commute; body or health sensitivity; leave option available |
| Public signal | THERMOMETER | x-vvv-x | Weather/work mismatch detected. Open? |
| Gated explanation | External weather risk matched private work/body context. |
| Human branches | Draft message; call supervisor; drink water; take leave; ignore; save trail |
| Trailstate path | x-vvv-x → q-vvv-p → n-vvv-n → r-vvv-r → 0-vvv-0 |
What this paper claims — and what it does not claim
Defensible claim
State-first signal grammar
StateLens defines a state-first signal grammar for situation-aware assistance: compact public operators, gated private context, human-confirmed branches, and replayable provenance.
Boundaries
Conceptual protocol paper
It is not a new foundation model, sensor system, medical device, emergency service or empirical effectiveness study. It is an interface/protocol pattern that still needs implementation and usability testing.
Earlier foundation: StateLens v1.0
The earlier StateLens v1.0 DOI remains part of the lineage. It should be cited when the topic is AI state diaries, multimodal state compression, URL-native operator trails, and day reconstruction.
StateLens: A URL-Native AI State Diary Protocol for Multimodal State Compression and Day Reconstruction
Eissens, R. (2026). StateLens: A URL-Native AI State Diary Protocol for Multimodal State Compression and Day Reconstruction (1.0). Zenodo. https://doi.org/10.5281/zenodo.20770792
World input → Multimodal AI → State resolution → URL-native operator trail → Diary reconstruction