Drop tensorflow deps; rule cleanups; repo-mem MCP; bump pumpingStation@6ab585b

- package.json: remove @tensorflow/tfjs and @tensorflow/tfjs-node.
  Monster's TF code was already stripped; the deps were stale and kept
  pulling a heavy native binary back into every install.
- .gitignore: ignore .repo-mem/ regenerable indexes and per-session
  .claude/*.lock runtime files.
- CLAUDE.md: prepend READ-FIRST pointer to .claude/rules/repo-mem.md;
  collapse the 'three outputs' bullet to a pointer at node-architecture.
- .claude/rules/telemetry.md: drop Port 0/1/2 duplication; reference
  node-architecture.md.
- .claude/rules/testing.md: stop requiring a separate test/edge tier and
  the basic/integration/edge example flow trio. Reflects what nodes
  actually do.
- .claude/rules/repo-mem.md (new): when-to-call-which guide for the
  per-repo memory MCP, anti-patterns, refresh model.
- .mcp.json (new): wire repo-mem stdio server.
- docs/DEVELOPER_GUIDE.md (new): step-by-step guide for adding a new
  EVOLV node under the three-layer pattern.
- Bump nodes/pumpingStation to 6ab585b (docs + simulations refresh,
  spill-flow path renames consistent with d8490aa).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.claude/rules/repo-mem.md

@@ -0,0 +1,80 @@
+# repo-mem MCP Tools
+
+This repo has a per-repo memory MCP server (`repo-mem`) wired via `.mcp.json`. It exposes 5 tools backed by a Hopfield substrate trained on EVOLV's source plus a BM25 index over file chunks. **Use them. They are faster and better-targeted than `grep` for concept queries, and they accumulate institutional memory of repairs.**
+
+If `/mcp` does not list `repo-mem` as Connected, the rest of this file does not apply for this session — fall back to `grep` / `Read`.
+
+## When to call which tool
+
+### `repo_search(query, k=8)` — primary lookup tool
+Use **before** `grep` / `find` / `Explore` agent for any natural-language "where is X handled / find all places that do Y / what code implements Z" question.
+
+- ✅ "where is the predicted volume integrator?" → `repo_search`
+- ✅ "find places that emit InfluxDB line protocol" → `repo_search`
+- ❌ "find every occurrence of `_updatePredictedVolume`" → `grep` (exact symbol — BM25 doesn't beat grep at exact-string lookup)
+- ❌ "list all `.test.js` files" → `find` / `ls` (no concept query)
+
+Returns top-K files with `file:line` ranges and snippets. Read the snippet first; only open the file if the snippet doesn't answer the question.
+
+### `repo_similar_fixes(query, failure?, files?, tags?, k=5)` — start-of-task context
+Call at the **start** of any non-trivial bug fix or behavioral change. Cheap (BM25 + file overlap + atom cosine), zero downside if it returns nothing useful.
+
+- Pass the user's task description as `query`.
+- If there's a failing test or stack trace, pass it as `failure`.
+- If you already know which files are involved, pass them as `files`.
+- Skim the returned traces; surface any near-match to the user before starting.
+
+### `repo_record_fix({task, failure, files, diff_summary, patch, tests, outcome, tags})` — end-of-task persist
+Call at the **end** of a landed fix or behavioral change, **before** reporting completion to the user. Skip for trivial typo/comment commits. Required fields: `task` and `outcome`. Recommended:
+- `failure`: the symptom that prompted the work (test output, user description, stack trace).
+- `files`: the files actually changed.
+- `diff_summary`: 1–3 sentences on *what* changed and *why*.
+- `patch`: the unified diff (truncate to the load-bearing hunks if huge).
+- `tests`: the verification command(s) you ran.
+- `outcome`: `passed` / `failed` / `partial` / `reverted`.
+- `tags`: short labels (`overflow-clamp`, `tokenizer`, `migration`, etc.) for retrieval bias.
+
+Rule of thumb: if the change took more than one read+edit pair, record it.
+
+### `substrate_score(text, worst_k=5)` — OOD-token check
+Use **sparingly**. After generating a non-trivial code block (≥ ~30 lines of new logic, not test scaffolding), pass it through `substrate_score` and inspect the worst-confidence positions for typos, wrong identifiers, or out-of-house style. Noisy on small additions — don't use it for one-line tweaks.
+
+### `substrate_top_next(context, k=10)` — rarely
+Predicts next BPE-subword tokens in the local style. Mostly useful for autonomous solver loops; in interactive review it's diagnostic only. If you find yourself wanting it, you probably want `repo_search` instead.
+
+## Workflow shape
+
+```
+new task arrives
+  ↓
+repo_similar_fixes(query=user_task)        ← cheap, always do this for non-trivial tasks
+  ↓
+repo_search(query=concept)                 ← when scoping
+  ↓
+[normal Read / Edit / Bash work]
+  ↓
+[after generating non-trivial new code]
+substrate_score(text=new_block)            ← optional, only if block is big
+  ↓
+[verify: tests / build / smoke]
+  ↓
+repo_record_fix({...})                     ← before final user-facing summary
+```
+
+## Anti-patterns
+
+- ❌ Calling `repo_search` when you already know the file path. Just `Read` it.
+- ❌ Calling `repo_record_fix` after every micro-edit. Only at meaningful task boundaries.
+- ❌ Treating `substrate_top_next` results as authoritative — they reflect repo style, not correctness.
+- ❌ Passing the full conversation to `substrate_score` — it's per-snippet, not per-session.
+
+## Refresh model
+
+The post-commit hook auto-runs `--quick --lock` (re-ingest + BM25 + chunk re-embed; substrate retrain skipped) so retrieval stays current within ~2 s of any commit. The substrate itself is only retrained when you (or a maintainer) run `--full` manually:
+
+```bash
+node ~/anchor-net-master/tools/repo-mem/refresh.mjs \
+  --repo . --in .repo-mem --full
+```
+
+Re-train when the repo gains substantially new vocabulary (new node, new domain, new dependency surface). Otherwise BM25 + existing atoms keep up.

.claude/rules/telemetry.md

@@ -5,10 +5,7 @@ paths:
 
 # Telemetry Rules
 
-## Output Port Convention
+Output port convention (Port 0/1/2) is documented in `.claude/rules/node-architecture.md`. This file covers only the Port 1 payload shape and downstream contracts.
-- Port 0: Process data (downstream node consumption)
-- Port 1: InfluxDB telemetry payload
-- Port 2: Registration/control plumbing
 
 ## InfluxDB Payload Structure
 Port 1 payloads must follow InfluxDB line protocol conventions:

.claude/rules/testing.md

@@ -5,18 +5,18 @@ paths:
 
 # Testing Rules
 
-## 3-Tier Test Structure
+## Test Structure
-Every node must have:
+Every node has at minimum:
-- `test/basic/*.test.js` — Unit tests for individual functions
+- `test/basic/*.test.js` — Unit tests for individual functions (specificClass domain logic)
 - `test/integration/*.test.js` — Node interaction and message passing tests
-- `test/edge/*.test.js` — Edge cases, error conditions, boundary values
 - `test/helpers/` (optional) — Shared test utilities for this node
 
+Edge-case tests live wherever they fit (in `basic/` for pure-logic edges, in `integration/` for runtime edges). Don't require a separate `test/edge/` directory.
+
 ## Test Runner
 ```bash
 node --test nodes/<nodeName>/test/basic/*.test.js
 node --test nodes/<nodeName>/test/integration/*.test.js
-node --test nodes/<nodeName>/test/edge/*.test.js
 ```
 
 ## Test Requirements
@@ -25,11 +25,7 @@ node --test nodes/<nodeName>/test/edge/*.test.js
 - Example flows (`examples/`) must stay in sync with implementation
 
 ## Example Flows
-Each node must maintain:
+Each node should ship at least one runnable example under `examples/` plus an `examples/README.md` describing it. Beyond that, add only what the node's complexity demands — not every node needs separate basic/integration/edge flow files.
-- `examples/README.md`
-- `examples/basic.flow.json`
-- `examples/integration.flow.json`
-- `examples/edge.flow.json`
 
 ## No Node-RED Runtime in Unit Tests
 Basic tests should test specificClass domain logic without requiring a running Node-RED instance.

.gitignore

@@ -8,3 +8,8 @@ npm-debug.log*
 
 # Build artifacts
 *.tgz
+# repo-mem regenerable indexes
+.repo-mem/
+
+# Per-session runtime locks (scheduled_tasks, etc.)
+.claude/*.lock

.mcp.json

@@ -0,0 +1,14 @@
+{
+  "mcpServers": {
+    "repo-mem": {
+      "type": "stdio",
+      "command": "node",
+      "args": [
+        "/home/znetsixe/anchor-net-master/tools/repo-mem/server.mjs",
+        "--in",
+        "/home/znetsixe/EVOLV/.repo-mem"
+      ],
+      "env": {}
+    }
+  }
+}

CLAUDE.md

@@ -1,5 +1,7 @@
 # EVOLV - Claude Code Project Guide
 
+> **READ FIRST, BEFORE ANY OTHER WORK:** `.claude/rules/repo-mem.md` — this repo has an MCP server (`repo-mem`) exposing a substrate-trained `repo_search` and a persistent fix-trace store. Use those instead of grep for concept queries, and record completed fixes via `repo_record_fix`. Triggers, anti-patterns, and refresh model are in that rule.
+
 ## What This Is
 Node-RED custom nodes package for wastewater treatment plant automation. Developed by Waterschap Brabantse Delta R&D team. Follows ISA-88 (S88) batch control standard.
 
@@ -23,7 +25,7 @@ Each node follows a three-layer pattern:
 - S88 color scheme: Area=#0f52a5, ProcessCell=#0c99d9, Unit=#50a8d9, Equipment=#86bbdd, ControlModule=#a9daee
 - Config JSON files in `generalFunctions/src/configs/` define defaults, types, enums per node
 - Tick loop runs at 1000ms intervals for time-based updates
-- Three outputs per node: [process, dbase, parent]
+- Output ports + 3-tier architecture: see `.claude/rules/node-architecture.md`
 - **Multi-tab demo flows**: see `.claude/rules/node-red-flow-layout.md` for the tab/link-channel/spacing rule set used by `examples/`
 
 ## Development Notes

docs/DEVELOPER_GUIDE.md

@@ -0,0 +1,602 @@
+# EVOLV Developer Guide: Creating a New Node
+
+This guide walks through creating a new EVOLV node from scratch, following the project's three-layer architecture pattern.
+
+## Prerequisites
+
+- **Node.js** (v18+)
+- **Node-RED** installed globally or as a dev dependency
+- Clone the repo and run `npm install` in the root (no build step required)
+- The `generalFunctions` submodule must be initialized (`git submodule update --init`)
+
+## Architecture Overview
+
+Every EVOLV node follows a **three-layer pattern**:
+
+| Layer | File | Responsibility |
+|-------|------|---------------|
+| 1 - Wrapper | `<name>.js` | Registers the node type with Node-RED, sets up HTTP endpoints for menus/config |
+| 2 - Node Adapter | `src/nodeClass.js` | Bridges Node-RED with domain logic: config loading, tick loop, input routing, lifecycle |
+| 3 - Domain Logic | `src/specificClass.js` | Pure business logic with no Node-RED dependencies |
+
+Plus a UI definition: `<name>.html` for the Node-RED editor.
+
+## Step-by-Step: Creating a New Node
+
+We will create a hypothetical `flowMeter` node as an example.
+
+### Step 1: Create Directory Structure
+
+```
+nodes/flowMeter/
+  flowMeter.js          # Layer 1 - wrapper
+  flowMeter.html        # UI definition
+  src/
+    nodeClass.js        # Layer 2 - node adapter
+    specificClass.js    # Layer 3 - domain logic
+  test/
+    specificClass.test.js
+```
+
+### Step 2: Write the Wrapper (`flowMeter.js`)
+
+The wrapper registers the node type with Node-RED and exposes HTTP endpoints for dynamic menus and config data.
+
+```js
+const nameOfNode = 'flowMeter';
+const nodeClass = require('./src/nodeClass.js');
+const { MenuManager, configManager } = require('generalFunctions');
+
+module.exports = function(RED) {
+  // Register the node type
+  RED.nodes.registerType(nameOfNode, function(config) {
+    RED.nodes.createNode(this, config);
+    this.nodeClass = new nodeClass(config, RED, this, nameOfNode);
+  });
+
+  // Menu endpoint (dynamic dropdowns in the editor UI)
+  const menuMgr = new MenuManager();
+  RED.httpAdmin.get(`/${nameOfNode}/menu.js`, (req, res) => {
+    try {
+      const script = menuMgr.createEndpoint(nameOfNode, ['asset', 'logger', 'position']);
+      res.type('application/javascript').send(script);
+    } catch (err) {
+      res.status(500).send(`// Error generating menu: ${err.message}`);
+    }
+  });
+
+  // Config data endpoint (exposes JSON config to the editor)
+  const cfgMgr = new configManager();
+  RED.httpAdmin.get(`/${nameOfNode}/configData.js`, (req, res) => {
+    try {
+      const script = cfgMgr.createEndpoint(nameOfNode);
+      res.type('application/javascript').send(script);
+    } catch (err) {
+      res.status(500).send(`// Error generating configData: ${err.message}`);
+    }
+  });
+};
+```
+
+Key points:
+- `nameOfNode` must match the file name, the `registerType` name, and the HTML `data-template-name`.
+- Menu categories (`['asset', 'logger', 'position']`) control which shared UI sections appear.
+- The config endpoint is optional if you do not need dynamic config in the editor.
+
+### Step 3: Write `nodeClass.js` (Node Adapter)
+
+This class bridges Node-RED's API with your domain logic.
+
+```js
+const { outputUtils, configManager } = require('generalFunctions');
+const Specific = require('./specificClass');
+
+class nodeClass {
+  constructor(uiConfig, RED, nodeInstance, nameOfNode) {
+    this.node = nodeInstance;
+    this.RED = RED;
+    this.name = nameOfNode;
+
+    this._loadConfig(uiConfig);
+    this._setupSpecificClass();
+    this._bindEvents();
+    this._registerChild();
+    this._startTickLoop();
+    this._attachInputHandler();
+    this._attachCloseHandler();
+  }
+
+  _loadConfig(uiConfig) {
+    const cfgMgr = new configManager();
+    this.defaultConfig = cfgMgr.getConfig(this.name);
+
+    // buildConfig merges base sections (general, asset, functionality)
+    // with node-specific domain config from the UI
+    this.config = cfgMgr.buildConfig(this.name, uiConfig, this.node.id, {
+      // Add domain-specific config sections here:
+      flowSettings: {
+        maxFlow: uiConfig.maxFlow,
+        pipeSize: uiConfig.pipeSize,
+      },
+    });
+
+    this._output = new outputUtils();
+  }
+
+  _setupSpecificClass() {
+    this.source = new Specific(this.config);
+    this.node.source = this.source;
+  }
+
+  _bindEvents() {
+    // Subscribe to domain events for Node-RED status display
+    this.source.emitter.on('flowUpdate', (val) => {
+      this.node.status({ fill: 'green', shape: 'dot', text: `${val} ${this.config.general.unit}` });
+    });
+  }
+
+  _registerChild() {
+    // Delayed to avoid Node-RED startup race conditions
+    setTimeout(() => {
+      this.node.send([
+        null,
+        null,
+        {
+          topic: 'registerChild',
+          payload: this.node.id,
+          positionVsParent: this.config?.functionality?.positionVsParent || 'atEquipment',
+          distance: this.config?.functionality?.distance || null,
+        },
+      ]);
+    }, 100);
+  }
+
+  _startTickLoop() {
+    setTimeout(() => {
+      this._tickInterval = setInterval(() => this._tick(), 1000);
+    }, 1000);
+  }
+
+  _tick() {
+    this.source.tick();
+    const raw = this.source.getOutput();
+    const processMsg = this._output.formatMsg(raw, this.config, 'process');
+    const influxMsg  = this._output.formatMsg(raw, this.config, 'influxdb');
+    this.node.send([processMsg, influxMsg]);
+  }
+
+  _attachInputHandler() {
+    this.node.on('input', (msg, send, done) => {
+      switch (msg.topic) {
+        case 'measurement':
+          if (typeof msg.payload === 'number') {
+            this.source.inputValue = parseFloat(msg.payload);
+          }
+          break;
+        // Add more input topics as needed
+      }
+      done();
+    });
+  }
+
+  _attachCloseHandler() {
+    this.node.on('close', (done) => {
+      clearInterval(this._tickInterval);
+      done();
+    });
+  }
+}
+
+module.exports = nodeClass;
+```
+
+Essential methods every `nodeClass` must implement:
+- `_loadConfig()` -- merges default JSON config with UI config via `configManager.buildConfig()`
+- `_setupSpecificClass()` -- instantiates the domain class
+- `_registerChild()` -- sends a `registerChild` message on output port 2 (parent)
+- `_startTickLoop()` -- drives periodic output at 1-second intervals
+- `_tick()` -- calls `source.getOutput()` and formats via `outputUtils.formatMsg()`
+- `_attachInputHandler()` -- routes incoming `msg.topic` to domain methods
+- `_attachCloseHandler()` -- clears timers on node removal
+
+### Step 4: Write `specificClass.js` (Domain Logic)
+
+This is pure JavaScript with no Node-RED dependencies.
+
+```js
+const EventEmitter = require('events');
+const { logger, configUtils, configManager, MeasurementContainer } = require('generalFunctions');
+
+class FlowMeter {
+  constructor(config = {}) {
+    this.emitter = new EventEmitter();
+    this.configManager = new configManager();
+    this.defaultConfig = this.configManager.getConfig('flowMeter');
+    this.configUtils = new configUtils(this.defaultConfig);
+    this.config = this.configUtils.initConfig(config);
+
+    this.logger = new logger(
+      this.config.general.logging.enabled,
+      this.config.general.logging.logLevel,
+      this.config.general.name
+    );
+
+    // MeasurementContainer stores typed/positioned values
+    this.measurements = new MeasurementContainer({
+      autoConvert: true,
+      windowSize: this.config.smoothing?.smoothWindow || 10,
+    });
+    this.measurements.setChildId(this.config.general.id);
+    this.measurements.setChildName(this.config.general.name);
+
+    // Domain state
+    this.currentFlow = 0;
+  }
+
+  tick() {
+    // Called every 1 second by nodeClass._tick()
+    this.calculateFlow();
+  }
+
+  calculateFlow() {
+    // Your domain logic here
+    const flow = this.currentFlow;
+
+    // Store in MeasurementContainer using the chainable API:
+    //   .type(measType).variant(variant).position(pos).value(val, timestamp, unit)
+    this.measurements
+      .type(this.config.asset.type)
+      .variant('measured')
+      .position(this.config.functionality.positionVsParent)
+      .value(flow, Date.now(), this.config.asset.unit);
+
+    this.emitter.emit('flowUpdate', flow);
+  }
+
+  getOutput() {
+    return {
+      flow: this.currentFlow,
+    };
+  }
+}
+
+module.exports = FlowMeter;
+```
+
+Key patterns:
+- Always create an `emitter` (EventEmitter) -- parents subscribe to child events through it.
+- Use `MeasurementContainer` for storing measurements. The chainable API follows the pattern: `measurements.type(t).variant(v).position(p).value(val, timestamp, unit)`.
+- Expose `tick()` and `getOutput()` for the node adapter to call.
+- Use `logger` instead of `console.log`.
+
+### Step 5: Write the HTML (UI Definition)
+
+```html
+<script src="/flowMeter/menu.js"></script>
+<script src="/flowMeter/configData.js"></script>
+
+<script>
+  RED.nodes.registerType("flowMeter", {
+    category: "EVOLV",
+    color: "#a9daee",          // S88 Control Module color
+    defaults: {
+      name: { value: "flowMeter" },
+      maxFlow: { value: 100, required: true },
+      pipeSize: { value: 0.1, required: true },
+      // Standard fields (asset, logger, position)
+      uuid: { value: "" },
+      supplier: { value: "" },
+      category: { value: "" },
+      assetType: { value: "" },
+      model: { value: "" },
+      unit: { value: "" },
+      enableLog: { value: false },
+      logLevel: { value: "error" },
+      positionVsParent: { value: "" },
+      positionIcon: { value: "" },
+    },
+    inputs: 1,
+    outputs: 3,
+    outputLabels: ["process", "dbase", "parent"],
+    icon: "font-awesome/fa-tachometer",
+    label: function() {
+      return this.name || "flowMeter";
+    },
+    oneditprepare: function() {
+      // Wait for shared menu system to initialize
+      const waitForMenuData = () => {
+        if (window.EVOLV?.nodes?.flowMeter?.initEditor) {
+          window.EVOLV.nodes.flowMeter.initEditor(this);
+        } else {
+          setTimeout(waitForMenuData, 50);
+        }
+      };
+      waitForMenuData();
+    },
+    oneditsave: function() {
+      if (window.EVOLV?.nodes?.flowMeter?.assetMenu?.saveEditor) {
+        window.EVOLV.nodes.flowMeter.assetMenu.saveEditor(this);
+      }
+      if (window.EVOLV?.nodes?.flowMeter?.loggerMenu?.saveEditor) {
+        window.EVOLV.nodes.flowMeter.loggerMenu.saveEditor(this);
+      }
+      if (window.EVOLV?.nodes?.flowMeter?.positionMenu?.saveEditor) {
+        window.EVOLV.nodes.flowMeter.positionMenu.saveEditor(this);
+      }
+    },
+  });
+</script>
+
+<script type="text/html" data-template-name="flowMeter">
+  <div class="form-row">
+    <label for="node-input-maxFlow"><i class="fa fa-arrows-v"></i> Max Flow</label>
+    <input type="number" id="node-input-maxFlow" placeholder="100" />
+  </div>
+  <div class="form-row">
+    <label for="node-input-pipeSize"><i class="fa fa-circle-o"></i> Pipe Size (m)</label>
+    <input type="number" id="node-input-pipeSize" placeholder="0.1" step="0.01" />
+  </div>
+  <!-- Shared UI sections injected by MenuManager -->
+  <div id="asset-fields-placeholder"></div>
+  <div id="logger-fields-placeholder"></div>
+  <div id="position-fields-placeholder"></div>
+</script>
+
+<script type="text/html" data-help-name="flowMeter">
+  <p><b>Flow Meter Node</b>: Measures and processes flow data.</p>
+</script>
+```
+
+**S88 color scheme** (pick based on your node's hierarchy level):
+
+| S88 Level | Color | Text Color |
+|-----------|-------|-----------|
+| Area | `#0f52a5` | white |
+| Process Cell | `#0c99d9` | white |
+| Unit | `#50a8d9` | black |
+| Equipment | `#86bbdd` | black |
+| Control Module | `#a9daee` | black |
+
+All nodes must have **3 outputs**: `[process, dbase, parent]`.
+
+### Step 6: Create Config JSON Schema
+
+Create `nodes/generalFunctions/src/configs/flowMeter.json`. This defines defaults and validation rules for every config property. The `configManager` reads this file by node name.
+
+```json
+{
+  "general": {
+    "name": {
+      "default": "FlowMeter",
+      "rules": { "type": "string", "description": "Human-readable name." }
+    },
+    "id": {
+      "default": null,
+      "rules": { "type": "string", "nullable": true }
+    },
+    "unit": {
+      "default": "m3/h",
+      "rules": { "type": "string" }
+    },
+    "logging": {
+      "logLevel": {
+        "default": "info",
+        "rules": {
+          "type": "enum",
+          "values": [
+            { "value": "debug" }, { "value": "info" },
+            { "value": "warn" }, { "value": "error" }
+          ]
+        }
+      },
+      "enabled": { "default": true, "rules": { "type": "boolean" } }
+    }
+  },
+  "functionality": {
+    "softwareType": { "default": "flowMeter", "rules": { "type": "string" } },
+    "role": { "default": "Sensor", "rules": { "type": "string" } },
+    "positionVsParent": {
+      "default": "atEquipment",
+      "rules": {
+        "type": "enum",
+        "values": [
+          { "value": "atEquipment" }, { "value": "upstream" }, { "value": "downstream" }
+        ]
+      }
+    }
+  },
+  "asset": {
+    "supplier": { "default": "Unknown", "rules": { "type": "string" } },
+    "category": { "default": "sensor", "rules": { "type": "string" } },
+    "type": { "default": "flow", "rules": { "type": "string" } },
+    "model": { "default": "Unknown", "rules": { "type": "string" } },
+    "unit": { "default": "m3/h", "rules": { "type": "string" } }
+  }
+}
+```
+
+Each property has a `default` value and a `rules` object specifying the type (`string`, `number`, `boolean`, `enum`, `object`), optional constraints (`min`, `max`, `nullable`), and a description.
+
+### Step 7: Register with `package.json`
+
+Add your node to the root `package.json` under `node-red.nodes`:
+
+```json
+{
+  "node-red": {
+    "nodes": {
+      "flowMeter": "nodes/flowMeter/flowMeter.js"
+    }
+  }
+}
+```
+
+Restart Node-RED to pick up the new node.
+
+### Step 8: Add Tests
+
+Create `nodes/flowMeter/test/specificClass.test.js`. Tests target Layer 3 (domain logic) directly, without Node-RED.
+
+```js
+const FlowMeter = require('../src/specificClass');
+
+function makeConfig(overrides = {}) {
+  const base = {
+    general: { name: 'TestFlow', id: 'test-1', logging: { enabled: false, logLevel: 'error' } },
+    functionality: { softwareType: 'flowMeter', role: 'sensor', positionVsParent: 'atEquipment' },
+    asset: { category: 'sensor', type: 'flow', model: 'test', supplier: 'Test', unit: 'm3/h' },
+  };
+  for (const key of Object.keys(overrides)) {
+    base[key] = typeof overrides[key] === 'object' ? { ...base[key], ...overrides[key] } : overrides[key];
+  }
+  return base;
+}
+
+describe('FlowMeter specificClass', () => {
+  it('should create an instance', () => {
+    const fm = new FlowMeter(makeConfig());
+    expect(fm).toBeDefined();
+  });
+
+  it('should return output with expected keys', () => {
+    const fm = new FlowMeter(makeConfig());
+    const out = fm.getOutput();
+    expect(out).toHaveProperty('flow');
+  });
+
+  it('tick() should not throw', () => {
+    const fm = new FlowMeter(makeConfig());
+    expect(() => fm.tick()).not.toThrow();
+  });
+});
+```
+
+Run tests with: `npm test` (uses Jest). The project also supports `node:test` for basic smoke tests.
+
+**Test organization conventions** (based on existing nodes):
+- `test/specificClass.test.js` -- unit tests for domain logic
+- `test/basic/*.test.js` -- structural/smoke tests (module loads, exports exist)
+- `test/edge/*.test.js` -- edge case and boundary tests
+- `test/integration/*.test.js` -- multi-component integration tests
+
+## Key APIs Reference
+
+### MeasurementContainer
+
+Chainable storage for typed, positioned measurements. Used by every domain class.
+
+```js
+const { MeasurementContainer } = require('generalFunctions');
+const mc = new MeasurementContainer({ autoConvert: true, windowSize: 10 });
+mc.setChildId('node-id');
+mc.setChildName('PT-001');
+
+// Store a value
+mc.type('pressure').variant('measured').position('upstream').value(3.5, Date.now(), 'bar');
+
+// Parents subscribe to events via mc.emitter
+mc.emitter.on('pressure.measured.upstream', (data) => { /* { value, unit, ... } */ });
+```
+
+The event name follows the pattern: `{type}.{variant}.{position}`.
+
+### configManager.buildConfig()
+
+Merges the JSON config schema defaults with UI-provided values. Called in `nodeClass._loadConfig()`.
+
+```js
+const { configManager } = require('generalFunctions');
+const cfgMgr = new configManager();
+const defaults = cfgMgr.getConfig('myNode');             // loads myNode.json
+const config = cfgMgr.buildConfig('myNode', uiConfig, nodeId, domainOverrides);
+```
+
+### POSITIONS
+
+Canonical position constants. Use these instead of hardcoded strings.
+
+```js
+const { POSITIONS } = require('generalFunctions');
+// POSITIONS.UPSTREAM      = 'upstream'
+// POSITIONS.DOWNSTREAM    = 'downstream'
+// POSITIONS.AT_EQUIPMENT  = 'atEquipment'
+// POSITIONS.DELTA         = 'delta'
+```
+
+### outputUtils.formatMsg()
+
+Formats raw output data into either `process` or `influxdb` messages. Only sends changed fields.
+
+```js
+const { outputUtils } = require('generalFunctions');
+const out = new outputUtils();
+const processMsg = out.formatMsg(rawData, config, 'process');
+const influxMsg  = out.formatMsg(rawData, config, 'influxdb');
+node.send([processMsg, influxMsg]);
+```
+
+### childRegistrationUtils
+
+Manages parent-child node relationships. Parents use this to accept child registrations.
+
+```js
+const { childRegistrationUtils } = require('generalFunctions');
+const regUtils = new childRegistrationUtils(this); // 'this' is the parent specificClass
+// Called when a child's registerChild message arrives:
+regUtils.registerChild(childSource, positionVsParent, distance);
+```
+
+The parent's `registerChild()` method subscribes to the child's `measurements.emitter` events for data propagation.
+
+## Common Patterns
+
+### Parent-Child Registration
+
+1. Child sends `{ topic: 'registerChild', payload: nodeId, positionVsParent }` on output port 2.
+2. Parent's `nodeClass._attachInputHandler()` catches `msg.topic === 'registerChild'`.
+3. Parent calls `childRegistrationUtils.registerChild(child, position)`.
+4. Parent subscribes to child's `measurements.emitter` events (e.g., `'flow.measured.downstream'`).
+5. When the child updates a measurement, the parent's listener fires and updates its own state.
+
+### Tick Loop
+
+Every node runs a 1-second tick loop that:
+1. Calls `source.tick()` to advance domain logic.
+2. Calls `source.getOutput()` for current state.
+3. Formats into `process` and `influxdb` messages via `outputUtils.formatMsg()`.
+4. Sends on ports 0 (process) and 1 (dbase).
+
+The tick loop starts with a 1-second delay to allow child registration to complete.
+
+### Three-Output Format
+
+All nodes send on three ports: `node.send([processMsg, influxMsg, parentMsg])`.
+
+| Port | Purpose | When |
+|------|---------|------|
+| 0 | Process data for downstream nodes | Every tick (if changed) |
+| 1 | InfluxDB line protocol for persistence | Every tick (if changed) |
+| 2 | Parent registration/control messages | On startup; on parent commands |
+
+### Event-Driven Communication
+
+Nodes communicate via `EventEmitter`, not Node-RED wires:
+- `measurements.emitter` fires `{type}.{variant}.{position}` events.
+- Parents listen to children's emitters after registration.
+- The `emitter` on the specificClass itself is used for internal state changes (e.g., updating Node-RED node status display).
+
+## Checklist
+
+Before submitting a new node, verify:
+
+- [ ] Three-layer structure: wrapper, nodeClass, specificClass
+- [ ] Config JSON in `generalFunctions/src/configs/<name>.json`
+- [ ] Registered in root `package.json` under `node-red.nodes`
+- [ ] HTML registers under category `'EVOLV'` with correct S88 color
+- [ ] Three outputs: `[process, dbase, parent]`
+- [ ] Uses `logger` (not `console.log`)
+- [ ] Uses `MeasurementContainer` for measurement storage
+- [ ] Uses `outputUtils.formatMsg()` for output formatting
+- [ ] Tick loop cleans up in `_attachCloseHandler()`
+- [ ] Tests exist for specificClass domain logic
+- [ ] Node-specific UI fields plus shared placeholders (asset, logger, position)

package.json

@@ -54,8 +54,6 @@
   "license": "SEE LICENSE",
   "dependencies": {
     "@flowfuse/node-red-dashboard": "^1.30.2",
-    "@tensorflow/tfjs": "^4.22.0",
-    "@tensorflow/tfjs-node": "^4.22.0",
     "generalFunctions": "file:nodes/generalFunctions",
     "mathjs": "^13.2.0"
   },