19 — EXTRACT

Transform complete verified instruments into the Phase-2 schema without losing evidence lineage.

NO TOOL Nothing implements this step

Nothing flows from this system into the legal ontology database. That is currently the correct behaviour, not a defect: no critical value has passed human review, so nothing is eligible for release.

Why it is held: Phase 2 extraction is on HOLD by owner decision. Building an extractor before REV produces adjudicated truth would create a path for unverified values to reach the database.

Deterministic Instrument path 0 archived step 19 of 26

Where this output sits

Source evidence›Candidate evidence›Human decision›Audit acceptance›Release›Production

Source evidence, candidate evidence, human decision, audit acceptance, release and production are six separate things. This step produces the highlighted one and nothing further.

Scope

Intended to turn a fully reviewed instrument unit into a canonical structured package with provenance and an explicit release or HOLD state, so that verified values — and only verified values — can reach the legal database.

This role is allowed to

Map confirmed spans to schema fields
Preserve null/HOLD

How it works

fed by 18 REV

Inputs

Complete instrument unitVER decisions for all critical fieldsExact source/CAP provenance

Deterministic Deterministic schema extractor; specialized model only under a separate approved contract.

No internal sequence is documented for this step — only the contract above and below it. Nothing was invented to fill the gap.

Must pass

All required fields trace to exact spans
Any unresolved critical value keeps the instrument HOLD

passes →

Outputs

Canonical instrument JSON/table packageField-level source spans and provenanceRelease or HOLD state

hands off to Phase 2 candidate environment · Audit and human review

fails →

HOLD

EXTRACT_UNSUPPORTED_VALUEEXTRACT_PROVENANCE_GAPEXTRACT_CRITICAL_HOLDEXTRACT_SCHEMA_FAILURE

Nothing continues on a failed gate. Uncertainty becomes an explicit HOLD, and no later step may read an unanswered item as an accepted one.

Non-scope — what this role does NOT own

Inventing field values
Releasing an unverified critical value
Writing legal truth without a separate approved policy

Explicitly forbidden

Fill missing facts
Export mismatched or unreadable critical values
Write directly into legal truth without the release gate

Dependencies and position

Starts: After instrument boundaries are complete and critical fields are resolved.
Previous step: 18 REV — HUMAN REVIEW
Next step: 20 PROTECTION MANIFEST
Hands off to: Phase 2 candidate environment, Audit and human review

Exact inputs

Complete instrument unit
VER decisions for all critical fields
Exact source/CAP provenance

Exact outputs

Canonical instrument JSON/table package
Field-level source spans and provenance
Release or HOLD state

Performer and AI/ML boundary

Performer: Deterministic schema extractor; specialized model only under a separate approved contract.
Class: Deterministic
AI boundary: Not required in V1.

Training information

Not a trained component. No model, no weights, no training corpus. Behaviour is fixed by code and configuration, and identical inputs must produce identical outputs.

Deterministic validation and acceptance gates

All required fields trace to exact spans
Any unresolved critical value keeps the instrument HOLD

HOLD and failure behaviour

Failure codes this step may emit:

EXTRACT_UNSUPPORTED_VALUE
EXTRACT_PROVENANCE_GAP
EXTRACT_CRITICAL_HOLD
EXTRACT_SCHEMA_FAILURE

Uncertainty becomes an explicit HOLD. Omission never converts uncertainty into acceptance, and no downstream step may treat an unanswered item as an accepted one.

Downstream handoff

Phase 2 candidate environment, Audit and human review

Active tool

NO ACTIVE TOOL — HOLD. No implementation was invented to fill this gap.

Actual performer today: None for the current issue. Only the retired Rev3 issue-4678 extractor is documented, and it belongs to the abandoned single-witness architecture.
Missing implementation: A deterministic schema extractor producing a canonical instrument JSON/table package with provenance and an explicit release or HOLD state.
Consequence: Nothing flows from this system into the legal ontology database. That is currently the correct behaviour, not a defect: no critical value has passed human review, so nothing is eligible for release.
Why it is on HOLD: Phase 2 extraction is on HOLD by owner decision. Building an extractor before REV produces adjudicated truth would create a path for unverified values to reach the database.

Recorded in tool/TOOL_STATUS.json.

Archived versions

No superseded versions recorded for this role.

Known limitations

No active tool, and that is currently correct rather than a defect: no critical value has passed human review, so nothing is eligible for release.
Building an extractor before REV produces adjudicated truth would create a path for unverified values to reach the database.
Phase 2 extraction is on HOLD by owner decision.

Security and privacy

Operates on frozen local artifacts. No credentials required. No data leaves the machine.

Cost behaviour

No provider calls. No metered cost. Compute is local.

Copying a tool into this repository does not authorise running it, retraining it, calling a model, or processing a new issue. No paid call may be made without the owner's explicit authorisation and a hard cost cap.

Provenance

Registry: registry/roles.json (schema marsoom.e2t.roles.v1)
Derivation: 28 legacy roles - VER - HUMAN + REV = 27 active steps
Legacy role id: EXTRACT
Legacy source SHA-256: 6f5d5d7ed8869e45307424c2193f95ffab84abc44a6a188f5c7f98c2a48ec64d
Generated: 2026-09-04T22:13:12.433Z

Feedback and decisions

Feedback is recorded per source and never merged into an invented consensus. Where sources disagree, both positions stand and the owner decides.

Source	Events
Naser / owner Final authority. Overrides every other source.	none recorded
Codex / orchestrator Architecture and sequencing.	none recorded
Builder / designer Implementation reality and constraints.	none recorded
Independent reviewer Adversarial review of claims.	none recorded
Auditor Evidence verification against artifacts.	none recorded
Human REV reviewer Page-level truth from the review site.	none recorded

Recorded events

No feedback events recorded yet. The ledger exists and is append-only: feedback/FEEDBACK_LEDGER.jsonl.

Editing feedback is not possible: a change is a new event whose supersedes names the one it replaces, and the original stays exactly as written.

Generated from registry/roles.json by tools/build-reports.mjs. Do not hand-edit — edit the registry and rebuild.
Original page pixels are the visual authority. CAP owns the exact captured text.