Reference
A first-pass reference for the terms that recur most often across the site. The entries
below are seeded from scanner outputs in packages/sire-scanner/output and
then curated against the live product, architecture, and tool pages.
Core Surfaces
Terms in this section are here because they recur across the site and carry a lot of the product's meaning. The coverage badges show how broadly the term appears in the scanner output.
The Boundary Gate is the final emission checkpoint. It reads the Claim Ledger result and decides whether the response is authorized, partial, narrative-only, blocked, or needs tighter specification.
Why It Matters
The Ledger can score claims, but the Gate is what turns those scores into enforcement. It keeps unsupported decision-grade claims from leaving the system as if they were verified.
Coverage derived from acronym-level scanner output
Go DeeperThe Claim Ledger decomposes a response into atomic claims and checks each one against governed evidence. It records what held up, what failed, and why.
Why It Matters
This is the core verification surface in Kenshiki. Without it, you only score whole responses and cannot explain which assertion broke under scrutiny.
Coverage derived from acronym-level scanner output
Go DeeperKadai is the caller-facing reasoning API. It orchestrates governed retrieval, generation, evaluation, and output-state assignment before returning a response.
Why It Matters
Kadai is how applications consume governed synthesis without talking directly to a model and inheriting all of that model's unchecked authority.
Coverage derived from acronym-level scanner output
Go DeeperKura is Kenshiki's governed evidence store. It ingests source material, preserves provenance and retrieval boundaries, and defines what downstream answers are allowed to rely on.
Why It Matters
If Kura is weak, everything downstream turns into retrieval theater. It is the boundary between source material and claims the system can actually defend.
Coverage derived from acronym-level scanner output
Go DeeperThe Prompt Compiler rewrites a loose request into a governed query before any generation layer sees it. It applies CFPO structure so evidence, policy, and output expectations stay mechanically enforceable.
Why It Matters
This is where Kenshiki constrains the question up front instead of hoping post-generation scoring can repair a sloppy prompt after the fact.
Coverage derived from acronym-level scanner output
Go DeeperGovernance Terms
Terms in this section are here because they recur across the site and carry a lot of the product's meaning. The coverage badges show how broadly the term appears in the scanner output.
The Admissibility Gate runs before retrieval to verify that the required approved evidence exists for the operation in question. It is separate from SIRE exclusion and separate from authorization.
Why It Matters
This gate lets Kenshiki fail closed when evidence is missing, superseded, or not yet approved. That is a different control problem from retrieval relevance or access control.
Coverage derived from acronym-level scanner output
Go DeeperCFPO stands for Content, Format, Policy, and Output. It is the structure the Prompt Compiler uses to organize evidence, constraints, and response expectations inside the compiled prompt.
Why It Matters
CFPO is one of the ways Kenshiki turns prompt design from taste into machinery. It makes the prompt contract disciplined enough to audit and reproduce.
Coverage derived from acronym-level scanner output
Go DeeperThe evidence boundary is the line around what the system is allowed to use as real support for an answer. Everything inside it is governed, attributable, and scoped. Everything outside it is not authority.
Why It Matters
This is the conceptual center of the site. Kenshiki exists to keep the model from quietly stepping beyond the evidence boundary and inventing certainty.
Coverage derived from acronym-level scanner output
Go DeeperGoverned evidence is source material that has been ingested, structured, scoped, and made admissible inside the Kenshiki pipeline. It is not just retrieved text; it is evidence the system is prepared to defend.
Why It Matters
Kenshiki's thesis depends on this distinction. The system does not treat every nearby chunk as authority just because it can retrieve it.
Coverage derived from acronym-level scanner output
Go DeeperAn output state is the explicit classification attached to every Kenshiki response, such as AUTHORIZED, PARTIAL, REQUIRES_SPEC, NARRATIVE_ONLY, or BLOCKED.
Why It Matters
Output states make uncertainty legible. The reader does not have to infer whether a fluent answer is safe to use because the system says so directly.
Coverage derived from acronym-level scanner output
Go DeeperAlso seen as relationship-based access control
ReBAC stands for relationship-based access control. In Kenshiki, it scopes evidence by caller identity and document relationship so the model only sees material the caller is authorized to use.
Why It Matters
A governed answer is not just about whether evidence exists. It is also about whether this caller is allowed to use that evidence in the first place.
Coverage derived from acronym-level scanner output
Go DeeperAlso seen as Subject, Included, Relevant, Excluded
SIRE stands for Subject, Included, Relevant, and Excluded. It is the identity metadata that governs what a source covers, how it connects to other sources, and what it must never be used to answer.
Why It Matters
SIRE gives Kenshiki a deterministic identity layer for evidence. It is one of the main reasons the system can reproduce retrieval decisions instead of treating retrieval as a fuzzy side effect.
Coverage derived from acronym-level scanner output
Go DeeperArchitecture Terms
Terms in this section are here because they recur across the site and carry a lot of the product's meaning. The coverage badges show how broadly the term appears in the scanner output.
The bounded-synthesis pipeline is the end-to-end flow from identity and compilation through retrieval, generation, evaluation, and emission. Every stage operates inside a governed boundary.
Why It Matters
This is the runtime contract behind Kenshiki's thesis. The system is not just retrieving documents or scoring answers; it is constraining synthesis step by step.
Coverage derived from acronym-level scanner output
Go DeeperThe control plane is the part of Kenshiki that governs what the generation layer may see, what claims it may emit, and what policies apply across the system.
Why It Matters
Kenshiki positions itself as a control plane rather than just another model wrapper. The term explains where authority lives in the architecture.
Coverage derived from acronym-level scanner output
Go DeeperCross-plane policy propagation means governance rules defined in the control plane keep applying as requests move through build and orchestration layers instead of stopping at one integration boundary.
Why It Matters
This is Kenshiki's argument against safety theater. Policy is supposed to survive the whole trip from evidence selection to final emission.
Coverage derived from acronym-level scanner output
Go DeeperPortable agent identity is Kenshiki's idea that the identity and scope rules attached to evidence should travel with the runtime rather than being reinvented at each layer or deployment tier.
Why It Matters
This is part of how Kenshiki keeps the Workshop, Refinery, and Clean Room contract recognizable even as the proof boundary changes.
Coverage derived from acronym-level scanner output
Go DeeperKenshiki describes itself as a three-plane architecture: build, orchestration, and control. Build shapes evidence and prompts, orchestration runs inference, and control verifies and gates what leaves.
Why It Matters
The phrase explains how Kenshiki separates concerns without leaving governance stranded as a bolt-on afterthought.
Coverage derived from acronym-level scanner output
Go Deeper