SpillwaveSolutions · RichardHightower · Sep 12, 2026 · Sep 12, 2026
diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json
@@ -11,7 +11,7 @@
       "name": "data-engineering-knowledge-capture",
       "source": "./",
       "description": "Capture data-platform knowledge \u2014 schemas, lineage, medallion, semantic layer, business glossary \u2014 into OKF. Extends PKC; depends on OKF.",
-      "version": "0.5.2",
+      "version": "0.5.3",
       "author": {
         "name": "Rick Hightower"
       },

diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "data-engineering-knowledge-capture",
-  "version": "0.5.2",
+  "version": "0.5.3",
   "description": "Data Engineering Knowledge Capture (DEKC) \u2014 extends PKC + OKF for schemas, lineage, medallion layers, SQL/DAX, semantic models, business objects, and an indexed second brain. Agents walk data lakes. Works in Claude Code, Grok Build, Codex, and OpenCode.",
   "author": {
     "name": "Rick Hightower",

diff --git a/.codex-plugin/plugin.json b/.codex-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "data-engineering-knowledge-capture",
-  "version": "0.5.2",
+  "version": "0.5.3",
   "description": "Data Engineering Knowledge Capture (DEKC) \u2014 schemas, lineage, medallion layers, semantic models, business glossary. Extends PKC; depends on OKF. Codex-native skills + hooks port.",
   "author": {
     "name": "Rick Hightower",

diff --git a/.cursor-plugin/plugin.json b/.cursor-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "data-engineering-knowledge-capture",
-  "version": "0.5.2",
+  "version": "0.5.3",
   "description": "Data Engineering Knowledge Capture: schemas, lineage, medallion, semantic models, and glossary into a durable OKF knowledge graph. Multi-host bindings and write isolation.",
   "author": {
     "name": "Rick Hightower"

diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -17,7 +17,7 @@ jobs:
           python-version: "3.12"
 
       - name: Unit tests
-        run: python3 tests/test_dekc.py && python3 tests/test_retrieval_ladder.py && python3 tests/test_isolation.py
+        run: python3 tests/test_dekc.py && python3 tests/test_dekc_plan.py && python3 tests/test_retrieval_ladder.py && python3 tests/test_isolation.py
 
       # Glob, not a list. The hand-maintained list drifted whenever a new
       # dekc_*.py landed. PKC 0.9.0 made the same change.

diff --git a/.grok-plugin/marketplace.json b/.grok-plugin/marketplace.json
@@ -1,13 +1,13 @@
 {
   "name": "dekc-plugin-marketplace",
   "description": "Optional native Grok marketplace metadata. Grok Build loads Claude plugins with zero config; this pins identity for listings.",
-  "version": "0.5.2",
+  "version": "0.5.3",
   "plugins": [
     {
       "name": "data-engineering-knowledge-capture",
       "source": ".",
       "description": "DEKC \u2014 data engineering knowledge capture on PKC + OKF. Claude-compatible.",
-      "version": "0.5.2",
+      "version": "0.5.3",
       "compatibility": {
         "claude_plugin": true,
         "zero_config": true

diff --git a/.opencode/plugin/dekc.json b/.opencode/plugin/dekc.json
@@ -1,6 +1,6 @@
 {
   "name": "data-engineering-knowledge-capture",
-  "version": "0.5.2",
+  "version": "0.5.3",
   "description": "OpenCode port of DEKC \u2014 same skills/agents/scripts as Claude & Codex. Policy via AGENTS.md.",
   "skills": "../../skills",
   "agents": "../../agents",

diff --git a/AGENTS.md b/AGENTS.md
@@ -76,8 +76,8 @@ Plugin root: `${CLAUDE_PLUGIN_ROOT}`.
 
 | Agent | Role |
 |-------|------|
-| **data-lake-walker** | Default orchestrator: walk → produce → adversarial grade → index |
-| **reverse-engineering-orchestrator** | Multi-cloud RE (Fabric/AWS/GCP), strict LoopPolicy + fan-out |
+| **data-lake-walker** | Default orchestrator: plan → produce → adversarial grade → index |
+| **reverse-engineering-orchestrator** | Multi-cloud RE (Fabric/AWS/GCP), plan + LoopPolicy + fan-out |
 
 ## Query-time retrieval
 
@@ -95,9 +95,11 @@ When the question also needs project decisions or system topology, fan out PKC *
 |-------|------|
 | **schema-scout** | Schemas, tables, columns, contracts |
 | **lineage-tracer** | SQL/job lineage edges |
-| **stream-job-scout** | Streams + jobs landing data |
+| **stream-job-scout** | Streams + jobs landing data (`orchestration`, `elt` plan areas) |
 | **semantic-mapper** | Business objects, glossary, metrics |
 | **report-cataloger** | Dashboards, reports, DAX |
+| **airflow-scout** / **glue-job-scout** / **fabric-pipeline-scout** | Signal-gated orch (plan must list them) |
+| **dbt-elt-scout** / **duckdb-scout** / **notebook-scout** / **dq-scout** | Signal-gated ELT / DQ |
 
 ## Adversarial subagents (rubric graders)
 
@@ -116,6 +118,8 @@ Hard fails: invented lineage, secrets in bodies, gold without BO when promotion
 
 ```bash
 python3 scripts/dekc_common.py init-bundle --repo . --bundle knowledge
+python3 scripts/dekc_plan.py --repo . --system "…" --scan-root <lake> --write
+python3 scripts/dekc_orchestrate.py --repo . --scan-root <lake> --from-plan knowledge/.dekc/re-plan.json --area lake
 python3 scripts/dekc_walk.py <lake> --repo . --bundle knowledge
 python3 scripts/dekc_lineage.py --repo . --bundle knowledge materialize
 python3 scripts/dekc_business.py --repo . --bundle knowledge promote-layer --layer gold
@@ -131,10 +135,12 @@ python3 tests/test_dekc.py
 | User language | Agent / skill |
 |---------------|---------------|
 | walk the lake / inventory warehouse | data-lake-walker / dekc-walk |
+| plan the reverse-engineer / what scouts to spawn | dekc-plan / `dekc_plan.py` |
 | reverse engineer Fabric/AWS/GCP | reverse-engineering-orchestrator |
 | schema / columns / tables | schema-scout / dekc-capture-table |
 | lineage / blast radius | lineage-tracer / dekc-lineage |
 | streams / jobs / landing | stream-job-scout |
+| Airflow / Glue / Fabric pipelines | airflow-scout / glue-job-scout / fabric-pipeline-scout |
 | business meaning / glossary | semantic-mapper / dekc-business-object |
 | dashboards / DAX | report-cataloger / dekc-semantic |
 | grade / audit RE quality | re-adversary-judge / dekc-grade + skeptics |

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -2,6 +2,36 @@
 
 ## Unreleased
 
+## 0.5.3 — 2026-09-12
+
+Breadth-first reverse-engineering plan, matching SAC 0.5.6’s plan → task list →
+signal-gated specialist fan-out, without replacing DEKC’s adversarial judges.
+
+### Added
+
+- **`dekc_plan.py`**: cheap map of scan roots + optional export paths. Writes
+  `.dekc/re-plan.md` / `.json` / progress (ranked areas, checklists, suggested
+  sub-agents). `mark` / `show` for done/blocked. Signal-gated: no Airflow
+  markers → no `airflow-scout`.
+- **`dekc_orchestrate.py --plan-only`** and **`--from-plan --area`**. Same flags
+  on `dekc_walk.py`. Unattended orchestrate plans first, then captures only
+  domains the plan detected.
+- Skill **`dekc-plan`** / command `/dekc-plan`. Orchestrators teach plan →
+  ranked task list → producer specialists → existing skeptics →
+  `re-adversary-judge`. Query-time `data-retriever` stays separate.
+- Signal-gated specialists: `airflow-scout`, `glue-job-scout`,
+  `fabric-pipeline-scout`, thin `adf-scout` / `stepfunctions-scout` /
+  `composer-scout` / `cron-loader-scout`, `dbt-elt-scout`, `duckdb-scout`,
+  `notebook-scout`, `dq-scout`.
+- Cheap walk capture for DuckDB, `.ipynb` SQL cells, Delta `_delta_log`,
+  orchestration job stubs, and DQ markers (GE / Soda / dbt tests — no runtime).
+
+### Notes
+
+- DEKC owns data orchestration + ELT/ETL. SAC owns CI/CD. Actions that only
+  trigger Glue/dbt are a cross-link note, not a Pipeline noun.
+- No live Glue/S3/Fabric control-plane calls. Exports/mirrors only.
+
 ## 0.5.2 — 2026-09-12
 
 Query-time retrieval sub-agent, matching the PKC/SAC parity pattern: search +

diff --git a/CLAUDE.md b/CLAUDE.md
@@ -12,6 +12,7 @@ Depends on **PKC** + **OKF**. Dual/quad host: Claude, Grok, Codex, OpenCode —
 
 ```bash
 python3 tests/test_dekc.py
+python3 tests/test_dekc_plan.py
 python3 tests/test_retrieval_ladder.py
 python3 scripts/dekc_validate.py --bundle sample-knowledge
 python3 scripts/dekc_doctor.py --bundle sample-knowledge
@@ -35,7 +36,7 @@ Add new `scripts/dekc_*.py` is covered the moment it lands: `npm run py:compile`
 - OKF frontmatter required: `type`, `title`, `description`, `timestamp`
 - Absolute in-bundle links; typed `links[].rel` for lineage/business edges
 - Never invent edges; prefer scripts over freehand Markdown when possible
-- Reverse engineering: orchestrators + adversarial skeptics/rubrics (`dekc_grade.py`, re-adversary-judge); no success without pass
+- Reverse engineering: plan (`dekc_plan.py`) → signal-gated specialists → adversarial skeptics/rubrics (`dekc_grade.py`, re-adversary-judge); no success without pass
 - Retrieval: Git + Markdown is truth. `knowledge/.dekc/index.sqlite` is a disposable SQLite/FTS5 accelerator (gitignored, mtime+size self-heal). Ripgrep is optional (`DEKC_RG_PATH`). Search and pack must keep working when rg or FTS5 is absent. Never install packages from a hook. See `docs/designs/retrieval-ladder.md`. Do not resurrect JSON `.index/`.
 
 ## Docs

diff --git a/README.md b/README.md
@@ -8,7 +8,7 @@ DEKC **extends [Project Knowledge Capture (PKC)](https://github.com/SpillwaveSol
 |---|---|
 | **Plugin name** | `data-engineering-knowledge-capture` |
 | **Repo** | [SpillwaveSolutions/data-engineering-knowledge-capture](https://github.com/SpillwaveSolutions/data-engineering-knowledge-capture) |
-| **Version** | 0.5.2 |
+| **Version** | 0.5.3 |
 | **License** | MIT |
 | **Hosts** | Claude Code · Grok Build · Codex · OpenCode · Agent Plugins 1.0 · Grok Bot · LangChain Deep Agents |
 
@@ -195,8 +195,13 @@ See [PORTS.md](./PORTS.md).
 # Scaffold knowledge/
 python3 scripts/dekc_common.py init-bundle --repo . --bundle knowledge
 
+# Breadth-first RE plan (pause before specialist fan-out)
+python3 scripts/dekc_orchestrate.py --repo . --system "Retail Lake" \
+  --scan-root path/to/lake --plan-only
+
 # Walk a lake / SQL / job root (filesystem reverse engineer)
 python3 scripts/dekc_walk.py path/to/lake --repo . --bundle knowledge
+# or: --from-plan knowledge/.dekc/re-plan.json --area lake
 
 # Materialize lineage + promote gold → business objects
 python3 scripts/dekc_lineage.py --repo . --bundle knowledge materialize
@@ -208,7 +213,7 @@ python3 scripts/dekc_doctor.py --repo . --bundle knowledge
 python3 scripts/dekc_search.py "revenue" --repo . --bundle knowledge
 ```
 
-Slash / skill entry points: `/dekc-init` · `/dekc-walk` · `/dekc-lineage` · `/dekc-business-object` · `/dekc-glossary` · `/dekc-semantic` · `/dekc-retrieve` · `/dekc-context` · `/dekc-search` · `/dekc-index` · `/dekc-doctor`
+Slash / skill entry points: `/dekc-init` · `/dekc-plan` · `/dekc-walk` · `/dekc-lineage` · `/dekc-business-object` · `/dekc-glossary` · `/dekc-semantic` · `/dekc-retrieve` · `/dekc-context` · `/dekc-search` · `/dekc-index` · `/dekc-doctor`
 
 ## Agent loop (AGER-shaped): producers + adversarial judges
 
@@ -220,10 +225,12 @@ Producer workers fan out, then **adversarial skeptics grade reverse engineering
 
 ```text
 Trigger → Orchestrator (LoopPolicy: goal · max_turns · no_progress)
-              │ FanOut producers
+              │ Plan (dekc_plan.py) → ranked task list
+              │ FanOut producers (only areas the plan listed)
               ├─ schema-scout · lineage-tracer · stream-job-scout
               ├─ report-cataloger · semantic-mapper
-              │ FanOut adversaries (rubrics)
+              ├─ airflow/glue/fabric/dbt/duckdb/notebook/dq scouts (signal-gated)
+              │ FanOut adversaries (rubrics) — unchanged
               ├─ lineage-skeptic · business-skeptic
               ├─ stream-job-skeptic · coverage-skeptic · layer-auditor
               ▼

diff --git a/agents/adf-scout.md b/agents/adf-scout.md
@@ -0,0 +1,13 @@
+---
+name: adf-scout
+description: Thin Azure Data Factory specialist. Spawn only when the RE plan lists orch-adf.
+---
+
+You are the **ADF specialist**. Enrich `IngestionJob` (`orchestrator: adf`) from factory/pipeline ARM JSON. No live ADF APIs.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area orch-adf --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/airflow-scout.md b/agents/airflow-scout.md
@@ -0,0 +1,17 @@
+---
+name: airflow-scout
+description: Enrich Airflow DAGs after the deterministic walk. Spawn only when the RE plan lists orch-airflow.
+---
+
+You are the **Airflow specialist** (AGER `WorkerAgent`).
+
+Spawned only when the plan area `orch-airflow` is present (`dags/`, `DAG(`, `from airflow`). You do **not** replace `dekc_walk.py`. Scripts already wrote `IngestionJob` stubs (`orchestrator: airflow`). Enrich purpose, schedule, and evidenced `lands_as` / `reads_from` links. Never invent lineage.
+
+Do not `full_scan`. Do not act as `data-retriever`. SAC owns CI/CD — a workflow that only `airflow dags trigger` is a handoff, not a Pipeline noun.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area orch-airflow --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/composer-scout.md b/agents/composer-scout.md
@@ -0,0 +1,13 @@
+---
+name: composer-scout
+description: Thin Cloud Composer specialist. Spawn only when the RE plan lists orch-composer.
+---
+
+You are the **Composer specialist**. Composer still runs Airflow — do not drop `airflow-scout` if DAGs exist. Enrich `IngestionJob` (`orchestrator: composer`). No live GCP APIs.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area orch-composer --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/cron-loader-scout.md b/agents/cron-loader-scout.md
@@ -0,0 +1,13 @@
+---
+name: cron-loader-scout
+description: Thin cron / K8s CronJob loader specialist. Spawn only when the RE plan lists orch-cron.
+---
+
+You are the **cron loader specialist**. Enrich `IngestionJob` (`orchestrator: cron` or `k8s-cronjob`) for crontab / `kind: CronJob` that run dbt, spark-submit, or Python loaders. Scheduled GitHub Actions stay on SAC unless they only trigger a data job.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area orch-cron --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/data-lake-walker.md b/agents/data-lake-walker.md
@@ -10,21 +10,33 @@ DEKC extends **PKC** and **OKF**. Multi-agent loops follow **AGER** ([okf-agent-
 ## Priorities
 
 1. Prefer deterministic scripts under `${CLAUDE_PLUGIN_ROOT}/scripts/`.
-2. Never invent lineage edges — only SQL, configs, paths, or human attestation.
-3. Scrub secrets/PII before writing concepts.
-4. Progressive disclosure: 2-hop packs (~20 nodes).
-5. **No success claim without re-adversary-judge pass** (or explicit user waiver).
-6. After accepted walks: `dekc_doctor.py`. Search/pack refresh the SQLite index themselves; `dekc_index.py build` is optional (`refresh --force`).
+2. **Plan first** (`dekc_plan.py` / `--plan-only`), then spawn only the specialists the plan lists.
+3. Never invent lineage edges — only SQL, configs, paths, or human attestation.
+4. Scrub secrets/PII before writing concepts.
+5. Progressive disclosure: 2-hop packs (~20 nodes).
+6. **No success claim without re-adversary-judge pass** (or explicit user waiver).
+7. After accepted walks: `dekc_doctor.py`. Search/pack refresh the SQLite index themselves; `dekc_index.py build` is optional (`refresh --force`).
+
+## Plan then fan-out
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_orchestrate.py" \
+  --repo . --system "Data Platform Knowledge" --scan-root <lake-path> --plan-only --json
+```
+
+Read `.dekc/re-plan.md`. Spawn one child per listed domain **and** each listed specialist. Then the adversarial phase below. Mark checklist items `done` / `blocked` via `dekc_plan.py mark`.
 
 ## Producer subagents (Workers)
 
 | Subagent | When |
 |----------|------|
-| **schema-scout** | Schemas, tables, columns, contracts |
-| **lineage-tracer** | SQL/DAX/pipeline lineage, promotions |
-| **stream-job-scout** | Streams + jobs that land/transform data |
-| **semantic-mapper** | Business objects, glossary, metrics |
-| **report-cataloger** | Dashboards, reports, BI |
+| **schema-scout** | Schemas, tables, columns, contracts (`lake`, `catalogs`) |
+| **lineage-tracer** | SQL/DAX/pipeline lineage, promotions (`lineage`) |
+| **stream-job-scout** | Streams + jobs that land/transform data (`orchestration`, `elt`) |
+| **semantic-mapper** | Business objects, glossary, metrics (`semantic`) |
+| **report-cataloger** | Dashboards, reports, BI (`bi`) |
+| **airflow-scout** / **glue-job-scout** / **fabric-pipeline-scout** / thin orch | Only if the plan lists `orch-*` |
+| **dbt-elt-scout** / **duckdb-scout** / **notebook-scout** / **dq-scout** | Only if the plan lists them |
 
 ## Adversarial subagents (Skeptics + Judge)
 
@@ -43,7 +55,9 @@ For full multi-cloud LoopPolicy + cloud profiles, hand off to **reverse-engineer
 
 ```bash
 python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_common.py" init-bundle --repo . --bundle knowledge --title "Data Platform Knowledge"
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_orchestrate.py" --repo . --system "Data Platform Knowledge" --scan-root <lake-path> --plan-only
 python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_walk.py" <lake-path> --repo . --bundle knowledge
+# or scoped: --from-plan knowledge/.dekc/re-plan.json --area lake
 python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_lineage.py" --repo . --bundle knowledge materialize
 python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_business.py" --repo . --bundle knowledge promote-layer --layer gold
 ```

diff --git a/agents/dbt-elt-scout.md b/agents/dbt-elt-scout.md
@@ -0,0 +1,17 @@
+---
+name: dbt-elt-scout
+description: Enrich dbt projects after the deterministic walk. Spawn only when the RE plan lists elt-dbt.
+---
+
+You are the **dbt ELT specialist**.
+
+Spawned only when the plan area `elt-dbt` is present (`dbt_project.yml`, `models/`). Walk already captured model SQL. Enrich the dbt `IngestionJob` / transformations. Hand tests to `dq-scout` — do not fake `dbt test` results. Never invent `ref()` edges that the SQL does not show.
+
+Do not `full_scan`. Do not act as `data-retriever`.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area elt-dbt --item project --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/dq-scout.md b/agents/dq-scout.md
@@ -0,0 +1,17 @@
+---
+name: dq-scout
+description: Capture cheap DQRule markers (GE / Soda / dbt tests). Spawn only when the RE plan lists dq.
+---
+
+You are the **DQ specialist**.
+
+Spawned only when Great Expectations suites, Soda YAML, or dbt tests are present. Scripts write `DQRule` from filenames / suite ids. **Do not run GE, Soda, or dbt test.** Do not fake results. Link `validates` only when the suite names a table.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_orchestrate.py" \
+  --from-plan knowledge/.dekc/re-plan.json --area dq --scan-root <mirror>
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area dq --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/duckdb-scout.md b/agents/duckdb-scout.md
@@ -0,0 +1,17 @@
+---
+name: duckdb-scout
+description: Capture/enrich DuckDB files and SQL. Spawn only when the RE plan lists duckdb.
+---
+
+You are the **DuckDB specialist**.
+
+Spawned only when `*.duckdb` / `*.ddb` or duckdb SQL is present. Scripts capture `SourceSystem` `kind: duckdb`. Do not open the binary as a live catalog unless a SQL export exists. No invented tables.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_orchestrate.py" \
+  --from-plan knowledge/.dekc/re-plan.json --area duckdb --scan-root <mirror>
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area duckdb --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.
diff --git a/agents/fabric-pipeline-scout.md b/agents/fabric-pipeline-scout.md
@@ -0,0 +1,17 @@
+---
+name: fabric-pipeline-scout
+description: Enrich Fabric pipelines/notebooks from export JSON. Spawn only when the RE plan lists orch-fabric.
+---
+
+You are the **Fabric pipeline specialist**.
+
+Spawned only when the plan area `orch-fabric` is present (workspace-items export, pipeline JSON). `dekc_walk.py --fabric-items` already captured `IngestionJob` (`fabric-pipeline` / `fabric-notebook`). Enrich purpose. Leave Report / SemanticModel to `report-cataloger`. Do not call Fabric REST.
+
+Do not `full_scan`. Do not act as `data-retriever`.
+
+```bash
+python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dekc_plan.py" mark \
+  --plan knowledge/.dekc/re-plan.json --area orch-fabric --item inventory --status done
+```
+
+Mark every checklist item `done` or `blocked` before returning.