Continuity Taxonomy
thecontinuitylayer.com/continuity-taxonomyThe framework this project reads the field through, published as a dated, versioned asset: one canonical definition, eight reading axes, and seven properties for evaluating a continuity claim.
Canonical definition
A continuity layer is the part of an AI system that carries forward governed state across sessions, tools, and models: what is currently true, what changed, why it changed, and which evidence supports it. It sits above memory and retrieval, and below agent execution, turning stored history into usable, auditable continuity.
Eight reading axes
The dimensions this project reads any continuity claim, system, or paper along.
Memory
01What is stored.
See: Memory systems, Personal AI memory
Context
02What is available right now, in this call.
See: Context engineering
State
03What persists — the current shape of a project or task.
See: Stateful agents, Long-running workflows
Provenance
04Where each piece of truth came from.
See: Provenance & audit, Organizational memory
Agents
05Who acts, and under what authority.
See: Agent infrastructure
Workflows
06How continuity survives a handoff — session to session, tool to tool, agent to agent.
See: Long-running workflows, MCP & tool continuity
Evaluation
07How any of this is measured, and against what standard.
See: Eval benchmarks, Papers
Infrastructure
08Where the layer actually lives — protocol, runtime, product, or process.
See: MCP & tool continuity, Commercial
Seven continuity properties
What a system must do to earn the word continuity — the test this project applies to a claim.
Persistence beyond a session
01Does state outlive the conversation that created it?
Update handling
02When a fact changes, is the old value superseded with a reason, or silently overwritten?
Temporal ordering
03Can the system say what was true at a given time, not just what's true now?
Disambiguation
04Can it tell apart two similar-looking facts or entities instead of merging them?
Reconstruction
05Can it rebuild the current shape of a project or task, not just retrieve fragments of it?
Model independence
06Does the same state work if the model underneath changes?
Operational usefulness
07Does it actually change what a person or agent can do, or is it a record nobody consults?
Version history
This taxonomy is versioned. When it changes, a new version supersedes the last in the open — the record below is never rewritten.
- v1.0July 2026
Initial publication — the canonical continuity-layer definition, eight reading axes, and seven continuity properties, fixed as a versioned asset.
Versioning this asset is the project practicing continuity on its own core framework — the same supersede-with-a-reason discipline it uses to evaluate everyone else (see the Constitution).
Signature
A SHA-256 over the canonical serialization of this asset — the definition, the eight axes, the seven properties, the version, and a fixed publication timestamp. It is reproducible from the text above, so any copy whose bytes drift from this digest is self-evidently altered.
- Algorithm
- SHA-256
- Published (UTC)
- SHA-256
- e3a25029d910499f5cedb634f961b6888f98a98309fbd6a625d3ff5173f0619e