ultra-builder-pro-cli 0.14.0 → 0.15.0

package/docs/WORKFLOW-LIFECYCLE.md

@@ -0,0 +1,266 @@
+# Ultra Builder Pro workflow lifecycle
+
+This document defines when Ultra writes authoritative state, how each workflow
+advances, what invalidates evidence, and which decisions remain with the active
+host and project owner. It is the operational companion to
+[`ARCHITECTURE.md`](./ARCHITECTURE.md) and the machine contracts under `spec/`.
+
+## 1. Authority boundaries
+
+| Surface | Owns | Must not own |
+|---|---|---|
+| Active host and owner | Product judgment, research content, implementation, review findings, release authorization | Durable workflow status or fabricated evidence digests |
+| Skill | One reusable procedure and its evidence requirements | Project state, static product doctrine, host-specific tool fiction |
+| MCP | IDs, transitions, task contracts, evidence references, output hashes, gate verdicts, next action | Research prose, code generation, memory, or model calls |
+| `.ultra/state.db` | Baseline, change, task, workflow, session, event, incident, projection, and evidence authority | External memory or code-graph payloads |
+| Generated JSON and Markdown | Read-only projections and durable artifacts | Independent lifecycle authority |
+| Hook | Compact DB-derived lifecycle observation, minimal event metadata, and projection protection | Generic engineering judgment, prompt/transcript capture, parallel lifecycle logs, or arbitrary edit blocking |
+
+Prompt input can supply facts, content paths, reasons, owner decisions, and
+evidence references. MCP reads the current repository and DB, validates those
+inputs, hashes outputs, and derives status. Prompt input cannot override a task's
+public seam, verification command, context references, workflow summary, gate
+verdict, output digest, or next action.
+
+Every change-bound research, plan, dev, test, review, and deliver run inherits
+the baseline id from the owning change workflow. A Prompt cannot switch that
+authority, and no new stage may start on an archived or cancelled change. Doctor
+recovers a legacy change without this link into an explicitly blocked workflow;
+it does not fabricate the missing provenance.
+
+Ultra does not select an MVP. Product scope and delivery posture are owner
+decisions. A reduced research mode is valid only for an explicitly bounded active
+change with a recorded selection reason; initial greenfield and brownfield
+baselines always use all seventeen research steps.
+
+## 2. How `ultra-init` classifies a repository
+
+`task.init_project` scans a bounded repository view while ignoring `.git`,
+`.ultra`, dependencies, virtual environments, generated build output, and other
+known non-source directories.
+
+`auto` classification is deterministic:
+
+- `brownfield` when delivered-system evidence exists: application source, tests,
+  deployment configuration, or persisted-state/schema evidence;
+- `greenfield` when none of those signals exists. A Git repository, README,
+  package manifest, workspace manifest, or starter documentation alone does not
+  prove delivered behavior;
+- `migrated` is never inferred from source. It is a compatibility state created
+  when older Ultra authority is preserved and upgraded. It requires a new
+  evidence-backed brownfield adoption before normal work.
+
+The scan also records bounded signal paths, detected verification commands,
+evidence-backed project type and technology signals, monorepo markers, candidate
+workspace roots, and whether the scan limit was reached. Explicit owner-supplied
+project type or stack metadata overrides detection without deleting the observed
+signals. In a monorepo, the owner selects the repository scope; baseline
+worktree digests and accepted dirty files are restricted to that scope.
+
+Classification and readiness are separate. Initialization immediately creates
+or resumes `.ultra/state.db`, installs only the required scaffold, records the
+classification, opens an `init` workflow, and opens `full` research for
+greenfield or `adoption` research for brownfield. Empty templates are not an
+approved baseline.
+
+## 3. Durable status machines
+
+### Baseline
+
+```text
+draft -> adopting -> ready -> superseded
+            |
+            -> blocked -> adopting
+```
+
+- `draft`: greenfield authority exists but has not been established.
+- `adopting`: brownfield or migrated evidence is being recorded.
+- `blocked`: current baseline evidence, verification, scope, or approval cannot
+  converge.
+- `ready`: research, evidence, repository snapshot, gaps, and approval converged.
+- `superseded`: preserved historical baseline after an explicitly authorized
+  replacement.
+
+`ready` is not trusted as a label alone. Every baseline/status/doctor/change gate
+revalidates its persisted fields, current files, research outputs, revision and
+worktree, known-red acceptance, blocking gaps, and approval provenance. An ordinary
+active change may downgrade only the expected drift created after its completed
+`bind-baseline` step to warnings; it cannot downgrade adoption, migration, missing
+evidence, or structural blockers. Approved incident break-glass is the only other
+execution exception and creates a mandatory reconciliation gap.
+
+### Change
+
+```text
+active <-> blocked -> ready -> archived
+   \--------------------------> cancelled
+```
+
+Only a healthy `ready` baseline may authorize a new ordinary change. An incident
+may bypass an unhealthy baseline only with a stored reason and approver; archive
+then creates a blocking baseline-reconciliation gap.
+
+### Task
+
+```text
+pending -> in_progress -> completed
+   |            |
+   v            v
+ blocked <------+
+   |
+   -> pending
+
+pending -> expanded -> completed
+```
+
+A task is executable only when its DB contract contains an observable outcome,
+traceability, slice kind, public seam, exact verification command, acceptance
+checks, context references, resolved documentation impact, owner, and valid
+dependencies.
+
+### Workflow run and step
+
+```text
+run:  active <-> blocked -> ready -> completed
+                              \----> cancelled
+
+step: pending -> in_progress -> completed
+         |            |
+         v            v
+       blocked ------> in_progress/completed
+```
+
+Required steps execute in definition order. A step that requires evidence or an
+output cannot complete without it. When the final required step completes, the
+run becomes `ready`; `workflow.complete` revalidates the entire stage before
+making it `completed`.
+
+## 4. Write and verification lifecycle
+
+| Stage | Authoritative writes | Completion verification | Next route |
+|---|---|---|---|
+| `init` | Baseline classification/scope, init run, research run, initialization event, projection job | DB/schema opens, scaffold/projection succeeds; baseline convergence later completes init | `research`, re-adoption, or `change` for an already-ready baseline |
+| `research` | Ordered step evidence, decisions, typed semantic records with source digests, one immutable report per selected step, output paths and SHA-256 digests | Every report has the required structure; semantic kinds/attributes/source anchors are valid and current; full/adoption synthesis binds the current three specifications and distillate; all digests remain current | `baseline.record`, then approval/convergence; bounded change research returns to its change |
+| baseline convergence | Spec/source/runtime refs, verification results, known failures, unknowns, gaps, branch/HEAD/worktree, approval | Full/adoption research complete; required discovery/product/architecture refs current; revision/worktree exact; blocking gaps resolved or explicitly accepted where allowed | `change` for every initial or daily outcome |
+| `change` | Complete Change Contract, profile/risk rationale, research disposition, intent/delta artifacts, linked change run, initial context snapshot | Blocking decisions resolved; selected change research current; current plan task set matches DB; task-bound context is ready; one next action is derived | selected `research`, then `plan`, or `dev` for an already approved current plan |
+| `plan` | Plan-step evidence, complete change-owned task rows, acceptance coverage matrix, change-bound `.ultra/execution-plan.json`, explicit plan approval | Every Change acceptance id is owned by a task; task trace targets resolve; quick has one task; ownership/dependencies valid; no cycles; exported topology and digest exactly match the current change task set | `dev` for the first dependency-ready task |
+| `dev` | Task/session transitions, step evidence, immutable implementation context, completion/review references | Task completed; no live session; task contract unchanged; starting context and review artifacts valid; completion commits remain ancestors of current HEAD | Next task, then aggregate `test` |
+| `test` | Immutable checking context and `.ultra/reports/tests/<workflow-id>.json` | Report change/task set, full HEAD, worktree digest, context digest, commands, public seam, blockers, and all acceptance/regression/integration/static/build/performance/security/recovery dimensions agree; every omission has a not-applicable rationale | `review` when passing, otherwise resume `dev`/`test` |
+| `review` | Immutable review context, `task`/`change`/`plan` mode, worker selection provenance, two specialist axes, coordinated summary | Current diff/HEAD and context match; selected/skipped workers match recorded rationale; both axes complete; P0/P1 and axis verdicts deterministically derive final verdict | `deliver` only from approved `change` mode, otherwise resume implementation |
+| `deliver` | Convergence context, verified learning resolutions, schema-validated semantic reconciliation manifest, change archive, durable release-authorization decision, immutable delivery report | Tasks/test/change-review current; every baseline update has an anchored before/after semantic record; `change.converge` derives evidence from DB and accepts no Prompt verdicts; archive and baseline transaction agree; external actions have evidence | Report completion or start the next `change` |
+| `status` | None | Reads doctor, baseline, workflows, breadcrumb, tasks, sessions, Git, and DB-referenced reports | One exact current route |
+| `doctor` | None by default; explicit repair writes only mechanical recovery state | Re-runs integrity, schema, workflow-output, projection, session, incident, archive, and install checks | Exact repair, restore/rebaseline decision, or workflow route |
+
+Every mutating MCP operation appends an event and enqueues projection work when
+the corresponding read-only view can change. Projection failure is recorded as
+a retryable incident; it does not silently convert a failed write into success.
+`task.append_event` accepts only non-authoritative observations. Lifecycle events are
+emitted by their owning DB mutation and never act as a substitute state transition.
+
+## 5. Context and prompt follow-through
+
+`change.context` compiles an immutable Context Manifest v2 snapshot. It combines:
+
+- the bound baseline/change/task identity;
+- a role and lifecycle gate;
+- DB-backed task outcome, public seam, verification command, and required refs;
+- local file hashes, current HEAD, branch, and worktree digest;
+- optional metadata-only external memory or graph references;
+- advisory file/token/context-share budgets;
+- readiness blockers and one DB-derived next action.
+
+Critical stage workflows bind a matching snapshot as a durable output:
+
+| Workflow | Required role | Required gate |
+|---|---|---|
+| `change` | `plan` or `implement` | `planning` or `implementation` |
+| `dev` | `implement` | `implementation` |
+| `test` | `check` | `verification` |
+| `review` | `review` | `review` |
+| `deliver` | `check` | `convergence` |
+
+The snapshot path and digest are then carried into test, review, and delivery
+reports. This prevents a later Prompt from claiming evidence gathered under a
+different context. File and token budgets warn; they do not block legitimate
+work. Missing authority, required refs, task contracts, outputs, approval, or
+current gate evidence do block.
+
+## 6. Invalidation rules
+
+- Changing a task contract invalidates contexts and gates that depended on the
+  former contract.
+- Changing required source/spec content invalidates the recorded output or
+  context digest.
+- Changing a research semantic source invalidates that workflow even when its immutable
+  step report remains unchanged.
+- Changing full HEAD or the worktree invalidates aggregate test, review, and
+  delivery evidence.
+- A completed dev run remains valid across later task commits when its bound
+  task contract is unchanged and its recorded commits remain ancestors of the
+  current HEAD.
+- Test, review, and delivery reports use workflow-id paths so a later run cannot
+  overwrite historical evidence.
+- `.ultra` workflow artifacts are excluded from application-worktree drift, but
+  their own recorded SHA-256 digests are still authoritative.
+- Generated `tasks.json` and task-context headers are projections. Manual edits
+  are overwritten and never repair DB authority.
+
+## 7. Entry flows
+
+### New project
+
+```text
+ultra-init -> full research (17 steps) -> baseline approval
+-> initial change + plan -> dev per task -> test -> review -> deliver
+```
+
+### Existing project
+
+```text
+ultra-init -> brownfield adoption research (17 steps)
+-> characterization and known-red evidence -> gap ledger -> owner approval
+-> first selected change -> plan -> dev -> test -> review -> deliver
+```
+
+Adoption records the system that exists. It does not force the owner to recreate
+market, persona, or roadmap claims that cannot be supported; those remain
+`Unknown` or explicit gaps while every semantic step is still processed.
+
+### Earlier Ultra project
+
+```text
+doctor/migrate with backup -> migrated compatibility authority
+-> explicit brownfield re-adoption -> full convergence -> normal change loop
+```
+
+Migration preserves prior files and rows but never carries old approval forward.
+
+### Daily maintenance
+
+```text
+ultra-status -> ultra-change -> plan if task contracts are absent
+-> dev -> test -> review -> deliver -> reconciled baseline
+```
+
+Quick changes shorten the artifact burden only where the machine contract allows;
+they do not skip task ownership, executable acceptance, current checking, review,
+or baseline reconciliation. Incidents add structured diagnosis and recovery
+evidence rather than bypassing them.
+
+## 8. Recovery and diagnosis
+
+Use `system.doctor` for project authority and `ubp --doctor` for installed host
+assets. Both are read-only unless an explicit project repair command is chosen.
+
+During an explicit repair, a legacy active change that predates durable workflows gets
+one blocked change run at `bind-baseline` with
+`LEGACY_CHANGE_PROVENANCE_REQUIRED`; repair does not invent its evidence. Rebind it to
+the re-adopted baseline and resume the same run before planning. Every plan must use
+the exact baseline id stored by its owning change run; only an approved break-glass
+incident may legitimately bind no baseline id.
+
+Resume the same blocked workflow and current step after resolving its blocker.
+Do not create parallel runs to escape state. Supported schema and projection
+repairs are backup-first. A corrupt DB is preserved and requires an explicit
+restore or rebaseline decision; neither action can manufacture research,
+verification, or owner approval.

package/hooks/README.md

@@ -14,12 +14,17 @@ advisory workflow conditions into refusal gates.
 | Before compact | `workflow_checkpoint.py` | Validate and atomically save a minimal workflow checkpoint |
 | After compact/resume | `workflow_resume.py` | Prefer the DB breadcrumb; restore the minimal file checkpoint only when no active change owns recovery |
 | Stop | `pre_stop_check.py` | Report an incomplete workflow boundary and allow stop |
-| Subagent lifecycle | `subagent_tracker.py` | Append lifecycle evidence under `.ultra/runtime/` |
+| Subagent lifecycle | `subagent_tracker.py` | Append minimal start/stop metadata to the authoritative DB event stream |
 
 The plugin deliberately does not capture prompts, tool observations, transcripts, summaries, or
 cross-session memory or code-graph content. Generic command blocking and post-edit policy are
 user/repository governance. Persistent Memory/graph content belongs to separately installed providers.
 
+`subagent_tracker.py` passes only agent id/type and the host session id to the bundled
+`hook-event.cjs` helper. The helper binds the current change/task and appends
+`subagent_started` or `subagent_stopped` to `state.db`; it never stores transcript
+paths, messages, or a parallel JSONL lifecycle log.
+
 `context_spine.py` is an imported library, not an eighth lifecycle registration.
 Registering it separately would duplicate hook execution. The checkpoint is a recovery artifact, not a second authority. A valid newer
 live workflow wins; a newer valid checkpoint may restore a missing or older

package/hooks/health_check.py

@@ -14,7 +14,7 @@ REQUIRED_TABLES = {
     "tasks", "events", "sessions", "schema_version", "migration_history",
     "telemetry", "specs_refs", "circuit_breaker", "changes", "artifacts",
     "context_snapshots", "spec_learning_candidates", "trace_links", "incidents", "projection_jobs",
-    "event_consumers",
+    "event_consumers", "workflow_runs", "workflow_steps",
 }
 
 
@@ -72,12 +72,21 @@ def inspect(root: Path) -> dict:
                 ]
                 if str(code).startswith("BASELINE_")
             ]
+            baseline = breadcrumb.get("baseline") if breadcrumb else None
+            baseline_is_ready = isinstance(baseline, dict) and baseline.get("status") == "ready"
+            baseline_status = (
+                "fail" if baseline_codes and baseline_is_ready
+                else "warning" if baseline_codes
+                else "pass"
+            )
             report["checks"]["baseline"] = {
-                "baseline": breadcrumb.get("baseline") if breadcrumb else None,
+                "baseline": baseline,
                 "blockers": baseline_codes,
                 "readiness": "blocked" if baseline_codes else "ready",
-                "status": "warning" if baseline_codes else "pass",
+                "status": baseline_status,
             }
+            if baseline_status == "fail":
+                issues.append("baseline")
 
             incidents = [
                 {"id": row[0], "code": row[1], "severity": row[2], "retryable": bool(row[3])}

package/hooks/subagent_tracker.py

@@ -1,69 +1,67 @@
 #!/usr/bin/env python3
-"""Subagent lifecycle tracker.
+"""Append minimal Ultra subagent lifecycle metadata to state.db events.
 
-Logs SubagentStart/Stop events to .ultra/runtime/subagent-log.jsonl (project-level)
-for debugging and cost analysis.
-
-Protocol fields:
-  SubagentStart: agent_id, agent_type, session_id
-  SubagentStop: agent_id, agent_type, session_id, agent_transcript_path,
-                last_assistant_message, stop_hook_active
-
-Usage:
-  python3 subagent_tracker.py start  # called by SubagentStart hook
-  python3 subagent_tracker.py stop   # called by SubagentStop hook
+The hook never stores prompts, messages, transcript paths, or a parallel JSONL log.
+It is advisory: a failed lifecycle event is reported on stderr but never blocks the
+host's subagent operation.
 """
 
 import json
+import os
+import shutil
+import subprocess
 import sys
-from datetime import datetime, timezone
 from pathlib import Path
 
-from context_spine import ContextSpineError, find_root, read_breadcrumb
+from context_spine import find_root
 
-MAX_LOG_LINES = 5000
 
+def helper_path() -> Path:
+    hook_root = Path(__file__).resolve().parent
+    installed = hook_root.parent / "runtime" / "hook-event.cjs"
+    if installed.is_file():
+        return installed
+    source = hook_root.parent / "mcp-server" / "hook-event.cjs"
+    if source.is_file():
+        return source
+    raise RuntimeError("bundled hook lifecycle helper is missing")
 
-def get_log_dir(hook_input: dict):
-    root = find_root(Path(hook_input.get("cwd") or Path.cwd()).resolve())
-    if root is None:
-        return None
-    try:
-        breadcrumb = read_breadcrumb(root)
-    except ContextSpineError as exc:
-        print(f"[subagent_tracker] cannot inspect Context Spine: {exc}", file=sys.stderr)
-        return None
-    if breadcrumb and breadcrumb.get("change_id"):
-        return root / ".ultra" / "runtime"
-    return None
 
+def node_binary() -> str:
+    configured = os.environ.get("UBP_NODE")
+    if configured:
+        return configured
+    discovered = shutil.which("node")
+    if discovered:
+        return discovered
+    raise RuntimeError("node executable is unavailable")
 
-def rotate_log(log_file: Path) -> None:
-    """Keep last MAX_LOG_LINES entries to prevent unbounded growth."""
-    try:
-        if not log_file.exists():
-            return
-        with open(log_file, "r", encoding="utf-8") as f:
-            lines = f.readlines()
-        if len(lines) <= MAX_LOG_LINES:
-            return
-        with open(log_file, "w", encoding="utf-8") as f:
-            f.writelines(lines[-MAX_LOG_LINES:])
-    except OSError as exc:
-        print(f"[subagent_tracker] log rotation failed: {exc}", file=sys.stderr)
-
-
-def main():
-    if len(sys.argv) < 2:
-        print(json.dumps({}))
-        return
 
-    action = sys.argv[1]
+def append_event(root: Path, action: str, hook_input: dict) -> None:
+    payload = {
+        "agent_id": hook_input.get("agent_id", "unknown"),
+        "agent_type": hook_input.get("agent_type", "unknown"),
+        "session_id": hook_input.get("session_id", ""),
+    }
+    result = subprocess.run(
+        [node_binary(), str(helper_path()), str(root.resolve()), action],
+        input=json.dumps(payload),
+        text=True,
+        capture_output=True,
+        check=False,
+        timeout=5,
+    )
+    if result.returncode != 0:
+        detail = result.stderr.strip() or f"exit {result.returncode}"
+        raise RuntimeError(detail)
+
+
+def main() -> None:
+    action = sys.argv[1] if len(sys.argv) > 1 else ""
     if action not in ("start", "stop"):
         print(json.dumps({}))
         return
 
-    # Read hook input from stdin
     try:
         raw = sys.stdin.read()
         hook_input = json.loads(raw) if raw.strip() else {}
@@ -73,35 +71,12 @@ def main():
         print(f"[subagent_tracker] invalid hook input: {exc}", file=sys.stderr)
         hook_input = {}
 
-    # Lazy init: avoid module-level subprocess
-    log_dir = get_log_dir(hook_input)
-    if log_dir is None:
-        print(json.dumps({}))
-        return
-    log_file = log_dir / "subagent-log.jsonl"
-    log_dir.mkdir(parents=True, exist_ok=True)
-
-    entry = {
-        "timestamp": datetime.now(timezone.utc).isoformat(),
-        "event": f"subagent_{action}",
-        "agent_id": hook_input.get("agent_id", "unknown"),
-        "agent_type": hook_input.get("agent_type", "unknown"),
-        "session_id": hook_input.get("session_id", ""),
-    }
-
-    # SubagentStop provides additional fields
-    if action == "stop":
-        transcript = hook_input.get("agent_transcript_path", "")
-        if transcript:
-            entry["agent_transcript_path"] = transcript
-
-    try:
-        with open(log_file, "a", encoding="utf-8") as f:
-            f.write(json.dumps(entry, ensure_ascii=False) + "\n")
-    except OSError as exc:
-        print(f"[subagent_tracker] event write failed: {exc}", file=sys.stderr)
-
-    rotate_log(log_file)
+    root = find_root(Path(hook_input.get("cwd") or Path.cwd()).resolve())
+    if root is not None:
+        try:
+            append_event(root, action, hook_input)
+        except (OSError, RuntimeError, subprocess.TimeoutExpired) as exc:
+            print(f"[subagent_tracker] lifecycle event unavailable: {exc}", file=sys.stderr)
 
     print(json.dumps({}))
 

package/mcp-server/hook-event.cjs

@@ -0,0 +1,34 @@
+#!/usr/bin/env node
+'use strict';
+
+const fs = require('node:fs');
+const path = require('node:path');
+
+const runtime = fs.existsSync(path.join(__dirname, 'index.cjs'))
+  ? require('./index.cjs')
+  : require('./server.cjs');
+
+function main() {
+  const rootDir = path.resolve(process.argv[2] || process.cwd());
+  const action = process.argv[3];
+  let hookInput = {};
+  const raw = fs.readFileSync(0, 'utf8');
+  if (raw.trim()) {
+    hookInput = JSON.parse(raw);
+    if (!hookInput || typeof hookInput !== 'object' || Array.isArray(hookInput)) {
+      throw new Error('hook input must be a JSON object');
+    }
+  }
+  const result = runtime.appendHookLifecycleEvent({ rootDir, action, hookInput });
+  process.stdout.write(`${JSON.stringify(result)}\n`);
+}
+
+if (require.main === module) {
+  try { main(); }
+  catch (error) {
+    process.stderr.write(`hook lifecycle event failed: ${error.message}\n`);
+    process.exit(1);
+  }
+}
+
+module.exports = { main };

package/mcp-server/lib/archive-journal.cjs

@@ -31,6 +31,9 @@ function assertIntent(intent, rootDir) {
     || intent.schema_version !== INTENT_VERSION
     || typeof intent.change_id !== 'string' || !intent.change_id
     || !Array.isArray(intent.baseline_updates)
+    || typeof intent.reconciliation_path !== 'string' || !intent.reconciliation_path
+    || !/^[0-9a-f]{64}$/.test(String(intent.reconciliation_digest || ''))
+    || !intent.reconciliation_manifest || typeof intent.reconciliation_manifest !== 'object'
     || typeof intent.summary !== 'string' || intent.summary.trim().length < 3) {
     throw new ArchiveJournalError('ARCHIVE_INTENT_INVALID', 'archive intent shape is invalid');
   }
@@ -46,6 +49,7 @@ function assertIntent(intent, rootDir) {
     throw new ArchiveJournalError('ARCHIVE_INTENT_INVALID', 'archive intent destination is not the change archive root');
   }
   for (const update of intent.baseline_updates) resolveInside(rootDir, update, 'baseline update');
+  resolveInside(rootDir, intent.reconciliation_path, 'reconciliation path');
   if (intent.previous_summary !== null && typeof intent.previous_summary !== 'string') {
     throw new ArchiveJournalError('ARCHIVE_INTENT_INVALID', 'previous_summary must be string or null');
   }
@@ -77,16 +81,21 @@ function renderSummary(change, summary, baselineUpdates, noBaselineChangeReason)
   ].join('\n');
 }
 
-function sameOperation(intent, { change, summary, baselineUpdates, noBaselineChangeReason }) {
+function sameOperation(intent, {
+  change, summary, baselineUpdates, noBaselineChangeReason, reconciliationPath, reconciliationDigest,
+}) {
   return intent.change_id === change.id
     && intent.source === change.artifact_root
     && intent.summary === summary.trim()
     && JSON.stringify(intent.baseline_updates) === JSON.stringify(baselineUpdates)
-    && (intent.no_baseline_change_reason || null) === (noBaselineChangeReason || null);
+    && (intent.no_baseline_change_reason || null) === (noBaselineChangeReason || null)
+    && intent.reconciliation_path === reconciliationPath
+    && intent.reconciliation_digest === reconciliationDigest;
 }
 
 function prepareArchiveMove({
-  rootDir, change, summary, baselineUpdates, noBaselineChangeReason, now = new Date(),
+  rootDir, change, summary, baselineUpdates, noBaselineChangeReason,
+  reconciliationPath, reconciliationDigest, reconciliationManifest, now = new Date(),
 }) {
   const date = now.toISOString().slice(0, 10);
   const source = resolveInside(rootDir, change.artifact_root, 'source');
@@ -98,7 +107,7 @@ function prepareArchiveMove({
   if (!fs.existsSync(source) && fs.existsSync(destinationIntentFile)) {
     const resumed = readIntentFile(destinationIntentFile, rootDir);
     if (!sameOperation(resumed.intent, {
-      change, summary, baselineUpdates, noBaselineChangeReason,
+      change, summary, baselineUpdates, noBaselineChangeReason, reconciliationPath, reconciliationDigest,
     })) {
       throw new ArchiveJournalError('ARCHIVE_INTENT_CONFLICT', 'existing archive intent differs from retry input');
     }
@@ -115,7 +124,7 @@ function prepareArchiveMove({
   if (fs.existsSync(sourceIntentFile)) {
     prepared = readIntentFile(sourceIntentFile, rootDir);
     if (!sameOperation(prepared.intent, {
-      change, summary, baselineUpdates, noBaselineChangeReason,
+      change, summary, baselineUpdates, noBaselineChangeReason, reconciliationPath, reconciliationDigest,
     })) {
       throw new ArchiveJournalError('ARCHIVE_INTENT_CONFLICT', 'prepared archive intent differs from retry input');
     }
@@ -130,6 +139,9 @@ function prepareArchiveMove({
       summary: summary.trim(),
       baseline_updates: baselineUpdates,
       no_baseline_change_reason: noBaselineChangeReason || null,
+      reconciliation_path: reconciliationPath,
+      reconciliation_digest: reconciliationDigest,
+      reconciliation_manifest: reconciliationManifest,
       previous_summary: previousSummary,
       created_at: now.toISOString(),
     };