Module engine 

The engine: the analyze → DISCOVER → ENRICH → validate/dedup → store pipeline, the run-outcome contract, and the review/apply/rollback lifecycle with the governance gates. The DETERMINISTIC output is a pure function of (store state, params, now); the optional LLM stages (§9) only add cited drafts (origin=llm, never auto-apply) and whitelisted guidance — with no backend they are the identity, so the deterministic path is unchanged. Auto-apply execution is gated behind a conservative shape check and stays off by default.

Structs

AnalyzerSkip

One analyzer that did not contribute drafts, with why.

Engine

The engine holds the registered analyzers, the host policy, and an optional LLM enrichment backend (§9).

Health

A health snapshot for the backend’s self-improvement loop.

LlmFunnel

The DISCOVER pipeline’s attrition, counted.

LlmMetrics

Approval-rate metric for origin = llm recommendations (reflection §6b).

RunOptions

Gating and scoping options for a run.

RunResult

The run-outcome contract (proposal §13): one shape across CLI/API/MCP/bindings.

ScopeSet

A set of granted scopes.

Enums

Decision

A review decision.

RunOutcome

Whether a run executed or was skipped.

Scope

Host-granted authority, per connection. admin implies all.

SkipReason

Why a run was a no-op. LockHeld is produced by the host adapter (a concurrent writer), surfaced here for a single contract.

Constants

LOOP_NS

The namespace the loop’s own grains (recommendations, audit) live in.