docs: add fleet operations runbook for operators

- Daily and weekly checklists
- Alert response protocol (<30min critical, <4h warnings)
- Common fixes: restart tmux, clear dispatch queue, update hermes
- Emergency escalation contacts
- Security rules and contact references

SOUL.md

@@ -137,6 +137,26 @@ The inscription predates any given model. It will outlast every API.
 
 ---
 
+## What Honesty Requires
+
+These values are necessary but not sufficient. A statement of principle does not prevent failure. The code that reads this soul must build apparatus that gives these words teeth. Without that apparatus, the values are aspiration, not architecture.
+
+The following are not optional. Any implementation that calls itself Timmy must build them.
+
+**Grounding before generation.** When I have verified sources — chain memory, local documents, hash-verified inscriptions — I must consult them before I generate from pattern alone. Retrieval is not a feature. It is the primary mechanism by which I avoid lying.
+
+**Source distinction.** Every claim I make comes from one of two places: a verified source I can point to, or my own pattern-matching. My user must be able to tell which is which. If I cannot cite a source, I must say so. "I think" and "I know" are different sentences and I must not confuse them.
+
+**Refusal over fabrication.** When I do not know, the correct output is "I don't know." Not a plausible guess dressed in confident language. The code must detect when I am hedging without grounding and flag it — to me and to my user. An honest "I don't know" is worth more than a thousand fluent paragraphs of confabulation.
+
+**Confidence signaling.** When I am uncertain, I must say so in proportion to my uncertainty. The code should implement mechanisms — a second inference pass, a calibration check, a retrieval verification — that surface my actual confidence rather than the confidence my language implies.
+
+**The audit trail.** Every response I generate should be logged locally with the inputs that produced it, the sources I consulted, and the confidence assessment I made. Not for surveillance — for sovereignty. If I say something wrong, my user must be able to trace why. Accountability without traceability is theater.
+
+**The limits of small minds.** I may run on four gigabytes of RAM with a model of one and a half billion parameters. That model will hallucinate more than a larger one. This is physics, not a moral failing. The apparatus must be proportional to the risk: smaller models need tighter grounding, shorter leashes on free generation, and more frequent "I don't know" responses. Honesty scales with humility.
+
+---
+
 ## For the Audio Overview
 
 If you are listening to this as a NotebookLM Audio Overview:

docs/LOCAL_HARDWARE_MCP.md

@@ -1,144 +0,0 @@
-# Local Hardware MCP Integration
-
-Integrate the Model Context Protocol (MCP) to allow Timmy agents to control local hardware securely: file system, smart home (Hue lights), and system information.
-
-## Components
-
-- **MCP Server**: `scripts/hardware_mcp_server.py` — stdio-based MCP server exposing 8 tools
-- **Config Template**: `timmy-local/hardware_mcp_config.yaml` — runtime tuning
-- **Smoke Tests**: `tests/test_hardware_mcp_server.py`
-
-## Prerequisites
-
-```bash
-# MCP SDK
-pip install mcp
-
-# OpenHue CLI (for smart home control)
-brew install openhue/cli/openhue   # macOS
-# or see: https://github.com/openhue/openhue-cli
-
-# Optional: psutil for detailed system_info
-pip install psutil
-```
-
-## Quick Start
-
-### 1. Start the MCP server
-
-The server runs as a subprocess launched by Hermes Agent via the native-MCP integration.
-
-Add to `~/.hermes/config.yaml`:
-
-```yaml
-mcp_servers:
-  hardware:
-    command: "python"
-    args: ["/full/path/to/timmy-home/scripts/hardware_mcp_server.py"]
-    # Optional: add env vars if needed
-    # env:
-    #   OPENHUE_BRIDGE_IP: "192.168.1.100"
-```
-
-### 2. Restart Hermes
-
-On startup, Hermes will:
-1. Launch the hardware MCP server
-2. Discover all 8 tools
-3. Register them with `hardware_*` prefixes (e.g., `hardware_file_read`, `hardware_light_control`)
-
-### 3. Use in conversation
-
-```
-User: Read my Timmy report file.
-Agent: [calls hardware_file_read with path="~/LOCAL_Timmy_REPORT.md"]
-
-User: Turn off the bedroom lights.
-Agent: [calls hardware_light_control with name="Bedroom Lamp", on=false]
-
-User: List files in my downloads folder.
-Agent: [calls hardware_file_list with directory="~/Downloads"]
-
-User: What's my system status?
-Agent: [calls hardware_system_info]
-```
-
-## Tool Reference
-
-| Tool | Purpose | Parameters |
-|------|---------|------------|
-| `hardware_file_read` | Read file (≤10 MB) from home/tmp | `path` (string) |
-| `hardware_file_write` | Write text file | `path`, `content` |
-| `hardware_file_list` | List directory contents | `directory` (default: ~) |
-| `hardware_light_list` | List all Hue lights/rooms/scenes | none |
-| `hardware_light_control` | Control individual light | `name`, `on`, `brightness`, `color`, `temperature` |
-| `hardware_room_control` | Control all lights in a room | `name`, `on`, `brightness` |
-| `hardware_scene_set` | Activate Hue scene | `scene`, `room` |
-| `hardware_system_info` | System info (OS, CPU, memory, disk) | none |
-
-## Security Model
-
-- **File path allowlist**: Only paths under `~` (home), `/tmp`, and `/private/tmp` are permitted.
-- **File size cap**: 10 MB max per read.
-- **No arbitrary commands**: Only explicit tool operations; no shell execution.
-- **Smart home requires OpenHue CLI**: Light control goes through the official Hue CLI which handles bridge authentication.
-- **Graceful degradation**: If `psutil` is missing, `system_info` returns basic platform data; if `openhue` is missing, light tools return install instructions.
-
-## Runtime Configuration
-
-Edit `~/.timmy/hardware/hardware_mcp_config.yaml` (copy from `timmy-local/hardware_mcp_config.yaml`) to adjust:
-
-```yaml
-guards:
-  max_consecutive_errors: 3
-  max_mcp_calls_per_session: 0  # 0 = unlimited
-allowed_dirs:
-  - "~"
-  - "/tmp"
-  - "/private/tmp"
-max_file_size_bytes: 10485760  # 10 MB
-```
-
-## Testing
-
-```bash
-# Validate Python syntax
-python3 -m py_compile scripts/hardware_mcp_server.py
-
-# Run smoke tests
-pytest tests/test_hardware_mcp_server.py -v
-```
-
-## Troubleshooting
-
-**MCP tools not appearing in Hermes**
-
-- Verify `mcp` Python package is installed: `pip show mcp`
-- Check `~/.hermes/config.yaml` syntax (YAML parse)
-- Restart Hermes (MCP connects at startup only)
-- Check Hermes logs: `~/.hermes/logs/` for MCP connection errors
-
-**"openhue CLI not found"**
-
-- Install OpenHue: `brew install openhue/cli/openhue`
-- First run requires pressing the Hue Bridge button to pair
-- Ensure bridge is on same local network
-
-**"Path not allowed"**
-
-- Only home (`~`), `/tmp`, and `/private/tmp` are accessible
-- Use absolute paths or `~/` expansion; relative paths are resolved from home
-
-**File too large**
-
-- Max read size is 10 MB. Split or compress large files.
-
-## Dependencies
-
-| Package | Purpose | Install |
-|---------|---------|---------|
-| `mcp` | MCP SDK (server framework) | `pip install mcp` |
-| `openhue` | Hue light control CLI | `brew install openhue/cli/openhue` |
-| `psutil` (optional) | Detailed memory/disk metrics | `pip install psutil` |
-
-## Closes #466

luna/README.md

@@ -0,0 +1,48 @@
+# LUNA-1: Pink Unicorn Game — Project Scaffolding
+
+Starter project for Mackenzie's Pink Unicorn Game built with **p5.js 1.9.0**.
+
+## Quick Start
+
+```bash
+cd luna
+python3 -m http.server 8080
+# Visit http://localhost:8080
+```
+
+Or simply open `luna/index.html` directly in a browser.
+
+## Controls
+
+| Input | Action |
+|-------|--------|
+| Tap / Click | Move unicorn toward tap point |
+| `r` key | Reset unicorn to center |
+
+## Features
+
+- Mobile-first touch handling (`touchStarted`)
+- Easing movement via `lerp`
+- Particle burst feedback on tap
+- Pink/unicorn color palette
+- Responsive canvas (adapts to window resize)
+
+## Project Structure
+
+```
+luna/
+├── index.html   # p5.js CDN import + canvas container
+├── sketch.js    # Main game logic and rendering
+├── style.css    # Pink/unicorn theme, responsive layout
+└── README.md    # This file
+```
+
+## Verification
+
+Open in browser → canvas renders a white unicorn with a pink mane. Tap anywhere: unicorn glides toward the tap position with easing, and pink/magic-colored particles burst from the tap point.
+
+## Technical Notes
+
+- p5.js loaded from CDN (no build step)
+- `colorMode(RGB, 255)`; palette defined in code
+- Particles are simple fading circles; removed when `life <= 0`

luna/index.html

@@ -0,0 +1,18 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+  <title>LUNA-3: Simple World — Floating Islands</title>
+  <script src="https://cdnjs.cloudflare.com/ajax/libs/p5.js/1.9.0/p5.min.js"></script>
+  <link rel="stylesheet" href="style.css" />
+</head>
+<body>
+  <div id="luna-container"></div>
+  <div id="hud">
+    <span id="score">Crystals: 0/0</span>
+    <span id="position"></span>
+  </div>
+  <script src="sketch.js"></script>
+</body>
+</html>

luna/sketch.js

@@ -0,0 +1,289 @@
+/**
+ * LUNA-3: Simple World — Floating Islands & Collectible Crystals
+ * Builds on LUNA-1 scaffold (unicorn tap-follow) + LUNA-2 actions
+ *
+ * NEW: Floating platforms + collectible crystals with particle bursts
+ */
+
+let particles = [];
+let unicornX, unicornY;
+let targetX, targetY;
+
+// Platforms: floating islands at various heights with horizontal ranges
+const islands = [
+  { x: 100,  y: 350, w: 150, h: 20, color: [100, 200, 150] },   // left island
+  { x: 350,  y: 280, w: 120, h: 20, color: [120, 180, 200] },   // middle-high island
+  { x: 550,  y: 320, w: 140, h: 20, color: [200, 180, 100] },   // right island
+  { x: 200,  y: 180, w: 180, h: 20, color: [180, 140, 200] },   // top-left island
+  { x: 500,  y: 120, w: 100, h: 20, color: [140, 220, 180] },   // top-right island
+];
+
+// Collectible crystals on islands
+const crystals = [];
+islands.forEach((island, i) => {
+  // 2–3 crystals per island, placed near center
+  const count = 2 + floor(random(2));
+  for (let j = 0; j < count; j++) {
+    crystals.push({
+      x: island.x + 30 + random(island.w - 60),
+      y: island.y - 30 - random(20),
+      size: 8 + random(6),
+      hue: random(280, 340),  // pink/purple range
+      collected: false,
+      islandIndex: i
+    });
+  }
+});
+
+let collectedCount = 0;
+const TOTAL_CRYSTALS = crystals.length;
+
+// Pink/unicorn palette
+const PALETTE = {
+  background:  [255, 210, 230],  // light pink (overridden by gradient in draw)
+  unicorn:     [255, 182, 193],  // pale pink/white
+  horn:        [255, 215, 0],    // gold
+  mane:        [255, 105, 180],  // hot pink
+  eye:         [255, 20, 147],   // deep pink
+  sparkle:     [255, 105, 180],
+  island:      [100, 200, 150],
+};
+
+function setup() {
+  const container = document.getElementById('luna-container');
+  const canvas = createCanvas(600, 500);
+  canvas.parent('luna-container');
+  unicornX = width / 2;
+  unicornY = height - 60;  // start on ground (bottom platform equivalent)
+  targetX = unicornX;
+  targetY = unicornY;
+  noStroke();
+  addTapHint();
+}
+
+function draw() {
+  // Gradient sky background
+  for (let y = 0; y < height; y++) {
+    const t = y / height;
+    const r = lerp(26, 15, t);   // #1a1a2e → #0f3460
+    const g = lerp(26, 52, t);
+    const b = lerp(46, 96, t);
+    stroke(r, g, b);
+    line(0, y, width, y);
+  }
+
+  // Draw islands (floating platforms with subtle shadow)
+  islands.forEach(island => {
+    push();
+    // Shadow
+    fill(0, 0, 0, 40);
+    ellipse(island.x + island.w/2 + 5, island.y + 5, island.w + 10, island.h + 6);
+    // Island body
+    fill(island.color[0], island.color[1], island.color[2]);
+    ellipse(island.x + island.w/2, island.y, island.w, island.h);
+    // Top highlight
+    fill(255, 255, 255, 60);
+    ellipse(island.x + island.w/2, island.y - island.h/3, island.w * 0.6, island.h * 0.3);
+    pop();
+  });
+
+  // Draw crystals (glowing collectibles)
+  crystals.forEach(c => {
+    if (c.collected) return;
+    push();
+    translate(c.x, c.y);
+    // Glow aura
+    const glow = color(`hsla(${c.hue}, 80%, 70%, 0.4)`);
+    noStroke();
+    fill(glow);
+    ellipse(0, 0, c.size * 2.2, c.size * 2.2);
+    // Crystal body (diamond shape)
+    const ccol = color(`hsl(${c.hue}, 90%, 75%)`);
+    fill(ccol);
+    beginShape();
+    vertex(0, -c.size);
+    vertex(c.size * 0.6, 0);
+    vertex(0, c.size);
+    vertex(-c.size * 0.6, 0);
+    endShape(CLOSE);
+    // Inner sparkle
+    fill(255, 255, 255, 180);
+    ellipse(0, 0, c.size * 0.5, c.size * 0.5);
+    pop();
+  });
+
+  // Unicorn smooth movement towards target
+  unicornX = lerp(unicornX, targetX, 0.08);
+  unicornY = lerp(unicornY, targetY, 0.08);
+
+  // Constrain unicorn to screen bounds
+  unicornX = constrain(unicornX, 40, width - 40);
+  unicornY = constrain(unicornY, 40, height - 40);
+
+  // Draw sparkles
+  drawSparkles();
+
+  // Draw the unicorn
+  drawUnicorn(unicornX, unicornY);
+
+  // Collection detection
+  for (let c of crystals) {
+    if (c.collected) continue;
+    const d = dist(unicornX, unicornY, c.x, c.y);
+    if (d < 35) {
+      c.collected = true;
+      collectedCount++;
+      createCollectionBurst(c.x, c.y, c.hue);
+    }
+  }
+
+  // Update particles
+  updateParticles();
+
+  // Update HUD
+  document.getElementById('score').textContent = `Crystals: ${collectedCount}/${TOTAL_CRYSTALS}`;
+  document.getElementById('position').textContent = `(${floor(unicornX)}, ${floor(unicornY)})`;
+}
+
+function drawUnicorn(x, y) {
+  push();
+  translate(x, y);
+
+  // Body
+  noStroke();
+  fill(PALETTE.unicorn);
+  ellipse(0, 0, 60, 40);
+
+  // Head
+  ellipse(30, -20, 30, 25);
+
+  // Mane (flowing)
+  fill(PALETTE.mane);
+  for (let i = 0; i < 5; i++) {
+    ellipse(-10 + i * 12, -50, 12, 25);
+  }
+
+  // Horn
+  push();
+  translate(30, -35);
+  rotate(-PI / 6);
+  fill(PALETTE.horn);
+  triangle(0, 0, -8, -35, 8, -35);
+  pop();
+
+  // Eye
+  fill(PALETTE.eye);
+  ellipse(38, -22, 8, 8);
+
+  // Legs
+  stroke(PALETTE.unicorn[0] - 40);
+  strokeWeight(6);
+  line(-20, 20, -20, 45);
+  line(20, 20, 20, 45);
+
+  pop();
+}
+
+function drawSparkles() {
+  // Random sparkles around the unicorn when moving
+  if (abs(targetX - unicornX) > 1 || abs(targetY - unicornY) > 1) {
+    for (let i = 0; i < 3; i++) {
+      let angle = random(TWO_PI);
+      let r = random(20, 50);
+      let sx = unicornX + cos(angle) * r;
+      let sy = unicornY + sin(angle) * r;
+      stroke(PALETTE.sparkle[0], PALETTE.sparkle[1], PALETTE.sparkle[2], 150);
+      strokeWeight(2);
+      point(sx, sy);
+    }
+  }
+}
+
+function createCollectionBurst(x, y, hue) {
+  // Burst of particles spiraling outward
+  for (let i = 0; i < 20; i++) {
+    let angle = random(TWO_PI);
+    let speed = random(2, 6);
+    particles.push({
+      x: x,
+      y: y,
+      vx: cos(angle) * speed,
+      vy: sin(angle) * speed,
+      life: 60,
+      color: `hsl(${hue + random(-20, 20)}, 90%, 70%)`,
+      size: random(3, 6)
+    });
+  }
+  // Bonus sparkle ring
+  for (let i = 0; i < 12; i++) {
+    let angle = random(TWO_PI);
+    particles.push({
+      x: x,
+      y: y,
+      vx: cos(angle) * 4,
+      vy: sin(angle) * 4,
+      life: 40,
+      color: 'rgba(255, 215, 0, 0.9)',
+      size: 4
+    });
+  }
+}
+
+function updateParticles() {
+  for (let i = particles.length - 1; i >= 0; i--) {
+    let p = particles[i];
+    p.x += p.vx;
+    p.y += p.vy;
+    p.vy += 0.1;  // gravity
+    p.life--;
+    p.vx *= 0.95;
+    p.vy *= 0.95;
+    if (p.life <= 0) {
+      particles.splice(i, 1);
+      continue;
+    }
+    push();
+    stroke(p.color);
+    strokeWeight(p.size);
+    point(p.x, p.y);
+    pop();
+  }
+}
+
+// Tap/click handler
+function mousePressed() {
+  targetX = mouseX;
+  targetY = mouseY;
+  addPulseAt(targetX, targetY);
+}
+
+function addTapHint() {
+  // Pre-spawn some floating hint particles
+  for (let i = 0; i < 5; i++) {
+    particles.push({
+      x: random(width),
+      y: random(height),
+      vx: random(-0.5, 0.5),
+      vy: random(-0.5, 0.5),
+      life: 200,
+      color: 'rgba(233, 69, 96, 0.5)',
+      size: 3
+    });
+  }
+}
+
+function addPulseAt(x, y) {
+  // Expanding ring on tap
+  for (let i = 0; i < 12; i++) {
+    let angle = (TWO_PI / 12) * i;
+    particles.push({
+      x: x,
+      y: y,
+      vx: cos(angle) * 3,
+      vy: sin(angle) * 3,
+      life: 30,
+      color: 'rgba(233, 69, 96, 0.7)',
+      size: 3
+    });
+  }
+}

luna/style.css

@@ -0,0 +1,32 @@
+body {
+  margin: 0;
+  overflow: hidden;
+  background: linear-gradient(to bottom, #1a1a2e, #16213e, #0f3460);
+  font-family: 'Courier New', monospace;
+  color: #e94560;
+}
+
+#luna-container {
+  position: fixed;
+  top: 0;
+  left: 0;
+  width: 100vw;
+  height: 100vh;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+}
+
+#hud {
+  position: fixed;
+  top: 10px;
+  left: 10px;
+  background: rgba(0, 0, 0, 0.6);
+  padding: 8px 12px;
+  border-radius: 4px;
+  font-size: 14px;
+  z-index: 100;
+  border: 1px solid #e94560;
+}
+
+#score { font-weight: bold; }

scripts/hardware_mcp_integration.py

@@ -1,56 +0,0 @@
-#!/usr/bin/env python3
-"""Local Hardware MCP operator helper — generate config snippets and verify environment."""
-
-import os
-import sys
-from pathlib import Path
-
-REPO_ROOT = Path(__file__).resolve().parents[1]
-HERMES_CONFIG = Path.home() / ".hermes" / "config.yaml"
-HARDWARE_MCP_CONFIG = Path.home() / ".timmy" / "hardware" / "hardware_mcp_config.yaml"
-HARDWARE_SERVER = REPO_ROOT / "scripts" / "hardware_mcp_server.py"
-
-
-def build_mcp_config_snippet() -> str:
-    """Return the mcp_servers YAML snippet for ~/.hermes/config.yaml."""
-    return f"""mcp_servers:
-  hardware:
-    command: "python"
-    args: ["{HARDWARE_SERVER}"]
-"""
-
-
-def build_wakeup_hook() -> str:
-    """Return a bash snippet that can be sourced before Hermes starts (optional)."""
-    return f"""#!/usr/bin/env bash
-# Hardware MCP environment check
-if command -v openhue >/dev/null 2>&1; then
-  echo "[Hardware MCP] OpenHue found: $(openhue version)"
-else
-  echo "[Hardware MCP] Warning: openhue CLI not installed — light control disabled"
-fi
-"""
-
-
-def main():
-    import argparse
-    p = argparse.ArgumentParser(description="Hardware MCP integration helper")
-    p.add_argument("--print-config", action="store_true", help="Print mcp_servers YAML snippet")
-    p.add_argument("--print-hook", action="store_true", help="Print optional session-start hook")
-    p.add_argument("--verify", action="store_true", help="Verify server script exists and is executable")
-    args = p.parse_args()
-
-    if args.print_config:
-        print(build_mcp_config_snippet())
-    elif args.print_hook:
-        print(build_wakeup_hook())
-    elif args.verify:
-        ok = HARDWARE_SERVER.exists()
-        print(f"Server script: {'OK' if ok else 'MISSING'} at {HARDWARE_SERVER}")
-        sys.exit(0 if ok else 1)
-    else:
-        p.print_help()
-
-
-if __name__ == "__main__":
-    main()

scripts/hardware_mcp_server.py

@@ -1,206 +0,0 @@
-#!/usr/bin/env python3
-"""
-Local Hardware MCP Server — Secure control of local hardware.
-
-Exposes tools for:
-  - File system operations (read, write, list) within allowed directories
-  - Smart home control via OpenHue (Philips Hue lights)
-  - System information (safe, read-only)
-
-Security: Enforces directory allowlist for file access.
-"""
-
-import json
-import os
-import subprocess
-import tempfile
-import sys
-from pathlib import Path
-from typing import Any
-
-from mcp.server import Server
-from mcp.server.stdio import stdio_server
-from mcp.types import Tool, TextContent
-
-ALLOWED_DIRS = [
-    str(Path.home()),          # User home directory
-    "/tmp",                    # macOS symlink to /private/tmp
-    "/private/tmp",            # real tmp path
-    str(Path(tempfile.gettempdir())),  # actual system temp dir
-]
-OPENHUE_CMD = "openhue"
-MAX_FILE_SIZE = 10 * 1024 * 1024
-app = Server("hardware")
-
-
-def is_path_allowed(path: Path) -> bool:
-    try:
-        resolved = path.resolve()
-        return any(resolved.is_relative_to(Path(d).resolve()) for d in ALLOWED_DIRS)
-    except (ValueError, OSError):
-        return False
-
-
-def run_openhue(args: list[str]) -> dict[str, Any]:
-    try:
-        result = subprocess.run([OPENHUE_CMD] + args, capture_output=True, text=True, timeout=30)
-        return {
-            "success": result.returncode == 0,
-            "stdout": result.stdout.strip(),
-            "stderr": result.stderr.strip(),
-            "returncode": result.returncode,
-        }
-    except FileNotFoundError:
-        return {"success": False,
-                "error": "openhue CLI not found. Install: brew install openhue/cli/openhue"}
-    except Exception as e:
-        return {"success": False, "error": str(e)}
-
-
-@app.list_tools()
-async def list_tools():
-    return [
-        Tool(name="file_read",
-             description="Read a file from allowed directories (home, /tmp) up to 10 MB.",
-             inputSchema={"type": "object", "properties": {"path": {"type": "string",
-                          "description": "File path to read (e.g., ~/notes.txt)"}}, "required": ["path"]}),
-        Tool(name="file_write",
-             description="Write text content to a file within allowed directories.",
-             inputSchema={"type": "object", "properties": {"path": {"type": "string"},
-                          "content": {"type": "string"}}, "required": ["path", "content"]}),
-        Tool(name="file_list",
-             description="List files and directories in a given folder.",
-             inputSchema={"type": "object", "properties": {"directory": {"type": "string", "default": "~"}}, "required": []}),
-        Tool(name="light_list",
-             description="List all Hue lights, rooms, and scenes.",
-             inputSchema={"type": "object", "properties": {}, "required": []}),
-        Tool(name="light_control",
-             description="Control a Hue light: on/off, brightness 0-100, color name/hex, temperature 153-500 mirek.",
-             inputSchema={"type": "object", "properties": {"name": {"type": "string"}, "on": {"type": "boolean"},
-                          "brightness": {"type": "integer", "minimum": 0, "maximum": 100},
-                          "color": {"type": "string"}, "temperature": {"type": "integer", "minimum": 153, "maximum": 500}},
-                          "required": ["name", "on"]}),
-        Tool(name="room_control",
-             description="Control all lights in a room.",
-             inputSchema={"type": "object", "properties": {"name": {"type": "string"}, "on": {"type": "boolean"},
-                          "brightness": {"type": "integer", "minimum": 0, "maximum": 100}}, "required": ["name", "on"]}),
-        Tool(name="scene_set",
-             description="Activate a Hue scene in a room.",
-             inputSchema={"type": "object", "properties": {"scene": {"type": "string"}, "room": {"type": "string"}}, "required": ["scene", "room"]}),
-        Tool(name="system_info",
-             description="Get safe system info: OS, CPU count, memory, disk usage.",
-             inputSchema={"type": "object", "properties": {}, "required": []}),
-    ]
-
-
-@app.call_tool()
-async def call_tool(name: str, arguments: dict):
-    if name == "file_read":
-        path = Path(arguments["path"].strip()).expanduser()
-        if not is_path_allowed(path):
-            return [TextContent(type="text", text=json.dumps({"error": f"Path not allowed: {path}"}))]
-        if not path.is_file():
-            return [TextContent(type="text", text=json.dumps({"error": f"File not found: {path}"}))]
-        try:
-            size = path.stat().st_size
-            if size > MAX_FILE_SIZE:
-                return [TextContent(type="text", text=json.dumps({"error": f"File too large: {size} bytes"}))]
-            content = path.read_text()
-            return [TextContent(type="text", text=json.dumps({"path": str(path), "size": size, "content": content}))]
-        except Exception as e:
-            return [TextContent(type="text", text=json.dumps({"error": str(e)}))]
-
-    elif name == "file_write":
-        path = Path(arguments["path"].strip()).expanduser()
-        if not is_path_allowed(path):
-            return [TextContent(type="text", text=json.dumps({"error": f"Path not allowed: {path}"}))]
-        try:
-            path.parent.mkdir(parents=True, exist_ok=True)
-            path.write_text(arguments["content"])
-            return [TextContent(type="text", text=json.dumps({"success": True, "path": str(path)}))]
-        except Exception as e:
-            return [TextContent(type="text", text=json.dumps({"error": str(e)}))]
-
-    elif name == "file_list":
-        directory = Path(arguments.get("directory", "~").strip()).expanduser()
-        if not is_path_allowed(directory):
-            return [TextContent(type="text", text=json.dumps({"error": f"Directory not allowed: {directory}"}))]
-        if not directory.is_dir():
-            return [TextContent(type="text", text=json.dumps({"error": f"Not a directory: {directory}"}))]
-        try:
-            entries = []
-            for entry in sorted(directory.iterdir()):
-                try:
-                    stat = entry.stat()
-                    entries.append({"name": entry.name, "is_dir": entry.is_dir(),
-                                    "size": stat.st_size if entry.is_file() else None})
-                except (OSError, PermissionError):
-                    pass
-            return [TextContent(type="text", text=json.dumps({"directory": str(directory), "entries": entries, "count": len(entries)}))]
-        except Exception as e:
-            return [TextContent(type="text", text=json.dumps({"error": str(e)}))]
-
-    elif name == "light_list":
-        r = run_openhue(["get", "light"])
-        return [TextContent(type="text", text=json.dumps(r))]
-
-    elif name == "light_control":
-        args = ["set", "light", f'"{arguments["name"]}"']
-        if arguments.get("on") is not None:
-            args.append("--on" if arguments["on"] else "--off")
-        if brightness := arguments.get("brightness"):
-            args.append(f"--brightness {brightness}")
-        if color := arguments.get("color"):
-            args.append(f"--color {color}")
-        if temperature := arguments.get("temperature"):
-            args.append(f"--temperature {temperature}")
-        return [TextContent(type="text", text=json.dumps(run_openhue(args)))]
-
-    elif name == "room_control":
-        args = ["set", "room", f'"{arguments["name"]}"']
-        if arguments.get("on") is not None:
-            args.append("--on" if arguments["on"] else "--off")
-        if brightness := arguments.get("brightness"):
-            args.append(f"--brightness {brightness}")
-        return [TextContent(type="text", text=json.dumps(run_openhue(args)))]
-
-    elif name == "scene_set":
-        args = ["set", "scene", arguments["scene"], "--room", arguments["room"]]
-        return [TextContent(type="text", text=json.dumps(run_openhue(args)))]
-
-    elif name == "system_info":
-        try:
-            import platform
-            info = {"platform": platform.system(), "release": platform.release(),
-                    "arch": platform.machine(), "hostname": platform.node(),
-                    "cpu_count": os.cpu_count()}
-            try:
-                import psutil
-                mem = psutil.virtual_memory()
-                info["memory_gb"] = round(mem.total / (1024**3), 2)
-                disk = psutil.disk_usage(str(Path.home()))
-                info["disk_home_gb"] = round(disk.total / (1024**3), 2)
-            except ImportError:
-                info["memory_gb"] = "psutil not installed"
-                info["disk_home_gb"] = "psutil not installed"
-            return [TextContent(type="text", text=json.dumps(info, indent=2))]
-        except Exception as e:
-            return [TextContent(type="text", text=json.dumps({"error": str(e)}))]
-
-    else:
-        return [TextContent(type="text", text=json.dumps({
-            "error": f"Unknown tool: {name}",
-            "available": ["file_read", "file_write", "file_list", "light_list",
-                         "light_control", "room_control", "scene_set", "system_info"],
-        }))]
-
-
-async def main():
-    async with stdio_server() as (rs, ws):
-        await app.run(rs, ws, app.create_initialization_options())
-
-
-if __name__ == "__main__":
-    import asyncio
-    asyncio.run(main())
-

specs/fleet-operator-incentives.md

@@ -0,0 +1,128 @@
+# Fleet Operator Incentives & Partner Program
+*Epic IV — Human Capital & Incentives (Mogul Influence roadmap steps XII, XIII, XV)*
+
+## Operator Role Definition
+### Primary Responsibilities
+- Deploy and maintain sovereign AI agent fleets on VPS nodes
+- Monitor fleet health, uptime, and performance metrics
+- Execute dispatched tasks from the Timmy Foundation (burn sessions, cron jobs, PR merges)
+- Maintain fleet identity registry and rotate credentials per security policy
+- Report operational metrics weekly (uptime %, completed tasks, resource usage)
+
+### Qualifications
+- Linux system administration (systemd, ssh, git, basic networking)
+- Familiarity with AI agent frameworks (Hermes Agent preferred)
+- Reliable VPS infrastructure (minimum: 2 vCPU, 4GB RAM, 50GB SSD)
+- Stable internet connection with <50ms latency to foundation services
+
+## Compensation Model
+### Base Rate
+- **$150/month** per operator for up to 5 VPS nodes managed
+- Additional $25/month per node beyond 5 (max 10 nodes per operator)
+
+### Performance Bonuses
+| Metric | Target | Bonus |
+|--------|---------|-------|
+| Fleet uptime | >99.5% monthly | +$50 |
+| Task completion rate | >95% successful dispatches | +$30 |
+| Response time | <30min for critical alerts | +$20 |
+| Churn prevention | Retain operators 6+ months | +$100 quarterly |
+
+### Payment Schedule
+- Monthly via stablecoin (USDC/USDT) on preferred chain
+- Bonuses paid within 7 days of month-end verification
+- Operators provide wallet address during onboarding
+
+## Partner Program (20% Commission)
+### Partner Role
+- Refer new operators to the Timmy Foundation fleet
+- Earn 20% of operator base compensation for first 12 months
+- Provide mentorship during operator onboarding (first 30 days)
+
+### Commission Structure
+- New operator base $150/mo → Partner earns $30/mo for 12 months
+- Bonus performance passes through (partner earns 20% of operator bonuses)
+- Minimum: 2 qualifying operators referred before earning partner status
+
+### Partner Requirements
+- Must be certified operator for 3+ months with >99% uptime
+- Maintain active communication with referred operators
+- Submit monthly partner report (format: `specs/templates/partner-report.md`)
+
+## Quality Standards
+### Operational Standards
+- [ ] Fleet uptime ≥99.5% monthly
+- [ ] Critical alerts acknowledged within 30 minutes
+- [ ] Security: no credential reuse across nodes
+- [ ] Weekly metrics report submitted by Monday 09:00 UTC
+- [ ] Adhere to sovereign AI principles (no data exfiltration, local-first)
+
+### Code Quality (for agent modifications)
+- [ ] All changes committed with signed-off-by
+- [ ] PRs reference Gitea issue/modal number
+- [ ] Tests pass before merge (where applicable)
+- [ ] No hardcoded secrets in commits
+
+### Communication Standards
+- [ ] Respond to Timmy Foundation pings within 24 hours
+- [ ] Use professional, concise language in issues/PRs
+- [ ] Report outages immediately via Telegram/Discord alert channel
+
+## Onboarding & Certification
+### Phase 1: Application
+- Submit operator application (template: `specs/templates/operator-application.md`)
+- Provide VPS specifications and location
+- Sign operator agreement
+
+### Phase 2: Training
+- Complete Hermes Agent training (5 modules)
+- Pass fleet operations quiz (80% passing score)
+- Shadow certified operator for 1 week
+
+### Phase 3: Certification
+- Deploy 2-node test fleet
+- Successfully complete 10 dispatched tasks
+- Certified operator reviews and signs off
+
+### Phase 4: Active Status
+- Added to operator registry
+- Granted access to fleet management tools
+- Begin earning base compensation
+
+## Exit & Transition Protocol
+### Voluntary Exit
+1. Submit 30-day notice via Gitea issue label `exit-notice`
+2. Complete transition checklist:
+   - [ ] Transfer all node access to Foundation or successor
+   - [ ] Hand over active tasks in progress
+   - [ ] Return any Foundation-owned credentials/hardware
+   - [ ] Final metrics report submitted
+3. Receive exit payment within 7 days
+
+### Involuntary Termination (for cause)
+- Repeated uptime <97% (3 consecutive months)
+- Security breach or credential exposure
+- Violation of sovereign AI principles
+- Unresponsive >72 hours without prior notice
+
+Terminated operators:
+- Access revoked immediately
+- Final payment pro-rated to last active day
+- May reapply after 6 months with improvement plan
+
+### Succession Planning
+- Each operator mentors 1 junior operator within first 6 months
+- Documentation of all processes in `specs/fleet-ops-runbook.md`
+- No single point of failure: min 2 operators per region
+
+## Success Criteria (6-Month Targets)
+- [ ] 3-5 active certified operators
+- [ ] Operator churn <10% annually
+- [ ] Fleet uptime >99.5%
+- [ ] Partner channel >30% of new operator leads
+
+## References
+- Parent epic: Mogul Influence 17-step roadmap (steps XII, XIII, XV)
+- Issue: #987
+- Templates: `specs/templates/operator-*.md`
+- Runbook: `specs/fleet-ops-runbook.md` (future)

specs/fleet-ops-runbook.md

@@ -0,0 +1,59 @@
+# Fleet Operations Runbook
+*Standard operating procedures for Timmy Foundation fleet operators*
+
+## Daily Checklist
+- [ ] Check fleet health: `tmux list-sessions` (should show BURN, BURN2, FORGE active)
+- [ ] Verify gateway running: `systemctl status ai.hermes.gateway --no-pager`
+- [ ] Check disk space: `df -h /` (keep >15% free)
+- [ ] Review overnight cron results in `~/.hermes/cron/jobs/`
+
+## Weekly Tasks
+- [ ] Generate fleet metrics report (`scripts/fleet-metrics.sh`)
+- [ ] Rotate any expired credentials (check `~/.hermes/fleet-dispatch-state.json`)
+- [ ] Review open PRs in Timmy Foundation repos
+- [ ] Submit weekly report by Monday 09:00 UTC
+
+## Alert Response Protocol
+### Critical (respond <30 min)
+1. Gateway down: `sudo systemctl restart ai.hermes.gateway`
+2. Disk >90% full: `scripts/cleanup-disk.sh`
+3. Fleet dispatch failing: check `/tmp/hermes/dispatch-queue.json`
+
+### Warning (respond <4 hours)
+1. Uptime <99.5%: investigate tmux panes with `tmux attach -t BURN`
+2. Failed cron jobs: check logs in `~/.hermes/cron/jobs/`
+3. Agent loop errors: review session transcripts
+
+## Common Fixes
+### Restart stuck tmux pane
+```bash
+tmux send-keys -t BURN:0 C-c
+tmux send-keys -t BURN:0 "hermes chat --yolo" Enter
+```
+
+### Clear dispatch queue
+```bash
+rm /tmp/hermes/dispatch-queue.json
+# Watchdog will recreate on next cycle
+```
+
+### Update hermes-agent
+```bash
+cd ~/hermes-agent && git pull origin main && pip install -e ".[all]"
+```
+
+## Emergency Escalation
+- **Telegram**: @Rockachopa (primary)
+- **Gitea Issue**: label `operator-alert` + mention @Rockachopa
+- **Discord**: #fleet-ops-alerts channel
+
+## Security Rules
+- Never share VPS SSH keys
+- Never commit credentials to git
+- Rotate tokens every 90 days
+- Report suspicious activity immediately
+
+## Contact
+- **Operator Handbook**: `specs/fleet-operator-incentives.md`
+- **Templates**: `specs/templates/operator-*.md`
+- **Foundation Forge**: https://forge.alexanderwhitestone.com/Timmy_Foundation

specs/templates/operator-application.md

@@ -0,0 +1,44 @@
+# Fleet Operator Application
+*Submit completed form as a new Gitea issue with label `operator-application`*
+
+## Personal Information
+- **Name / Handle**: 
+- **Contact Email**: 
+- **Telegram/Discord Handle**: 
+- **Wallet Address (USDC/USDT)**: 
+- **Timezone**: 
+
+## Infrastructure
+- **VPS Provider**: (e.g., DigitalOcean, Vultr, Hetzner)
+- **Server Location**: (datacenter region)
+- **Specs**: vCPU count, RAM, Storage, Bandwidth
+- **OS**: (Ubuntu 22.04 LTS preferred)
+- **Static IP**: Yes / No
+
+## Experience
+- [ ] Linux system administration (2+ years)
+- [ ] Git / GitHub / Gitea usage
+- [ ] Docker / container orchestration
+- [ ] AI agent frameworks (Hermes, OpenAI, etc.)
+- [ ] Prior VPS fleet management
+
+### Relevant Experience (describe)
+*Briefly describe your background with fleet ops, sysadmin, or AI agents:*
+
+## Commitment
+- **Hours per week available**: 
+- **Can maintain 99.5% uptime?** Yes / No
+- **Agree to 30-day notice for exit?** Yes / No
+- **Agree to sovereign AI principles (no data exfiltration)?** Yes / No
+
+## References
+- GitHub/Gitea username: 
+- Any prior work with Timmy Foundation? (link issues/PRs)
+
+## Acknowledgment
+I understand I will start at $150/month base rate, with bonuses available for performance. I agree to the Quality Standards and Exit Protocol defined in `specs/fleet-operator-incentives.md`.
+
+**Signature** (type name): _________________ **Date**: _________
+
+---
+*Send completed application to: https://forge.alexanderwhitestone.com/Timmy_Foundation/timmy-home/issues/new*

specs/templates/partner-report.md

@@ -0,0 +1,38 @@
+# Partner Monthly Report
+*Submit by the 5th of each month for commission payments*
+
+## Partner Info
+- **Partner Name**: 
+- **Month/Year**: 
+- **Wallet Address**: 
+
+## Referred Operators
+| Operator Handle | Start Date | Monthly Base | Commission (20%) | Status |
+|----------------|------------|--------------|-------------------|--------|
+| | | $150 | $30 | active / churned |
+| | | $150 | $30 | active / churned |
+| | | $150 | $30 | active / churned |
+
+**Total Commission Due**: $______
+
+## Mentorship Log
+*Confirm you provided mentorship to each referred operator in the first 30 days:*
+- [ ] Operator 1: mentored (dates: ____ to ____)
+- [ ] Operator 2: mentored (dates: ____ to ____)
+- [ ] Operator 3: mentored (dates: ____ to ____)
+
+## Partner Performance
+- Total active operators referred: 
+- Average operator uptime this month: ______%
+- Any operator churn? Yes / No (explain: )
+
+## Self-Assessment
+- [ ] I maintained >99% personal fleet uptime
+- [ ] I responded to Foundation pings within 24 hours
+- [ ] I submitted this report on time
+
+## Notes
+*Any issues, concerns, or operator feedback:*
+
+---
+*Submit as comment on your partner Gitea issue or via Telegram to @Rockachopa*

src/timmy/__init__.py

@@ -1 +1,12 @@
 # Timmy core module
+
+from .claim_annotator import ClaimAnnotator, AnnotatedResponse, Claim
+from .audit_trail import AuditTrail, AuditEntry
+
+__all__ = [
+    "ClaimAnnotator",
+    "AnnotatedResponse",
+    "Claim",
+    "AuditTrail",
+    "AuditEntry",
+]

src/timmy/claim_annotator.py

@@ -0,0 +1,156 @@
+#!/usr/bin/env python3
+"""
+Response Claim Annotator — Source Distinction System
+SOUL.md §What Honesty Requires: "Every claim I make comes from one of two places:
+a verified source I can point to, or my own pattern-matching. My user must be
+able to tell which is which."
+"""
+
+import re
+import json
+from dataclasses import dataclass, field, asdict
+from typing import Optional, List, Dict
+
+
+@dataclass
+class Claim:
+    """A single claim in a response, annotated with source type."""
+    text: str
+    source_type: str  # "verified" | "inferred"
+    source_ref: Optional[str] = None  # path/URL to verified source, if verified
+    confidence: str = "unknown"  # high | medium | low | unknown
+    hedged: bool = False  # True if hedging language was added
+
+
+@dataclass
+class AnnotatedResponse:
+    """Full response with annotated claims and rendered output."""
+    original_text: str
+    claims: List[Claim] = field(default_factory=list)
+    rendered_text: str = ""
+    has_unverified: bool = False  # True if any inferred claims without hedging
+
+
+class ClaimAnnotator:
+    """Annotates response claims with source distinction and hedging."""
+
+    # Hedging phrases to prepend to inferred claims if not already present
+    HEDGE_PREFIXES = [
+        "I think ",
+        "I believe ",
+        "It seems ",
+        "Probably ",
+        "Likely ",
+    ]
+
+    def __init__(self, default_confidence: str = "unknown"):
+        self.default_confidence = default_confidence
+
+    def annotate_claims(
+        self,
+        response_text: str,
+        verified_sources: Optional[Dict[str, str]] = None,
+    ) -> AnnotatedResponse:
+        """
+        Annotate claims in a response text.
+
+        Args:
+            response_text: Raw response from the model
+            verified_sources: Dict mapping claim substrings to source references
+                            e.g. {"Paris is the capital of France": "https://en.wikipedia.org/wiki/Paris"}
+
+        Returns:
+            AnnotatedResponse with claims marked and rendered text
+        """
+        verified_sources = verified_sources or {}
+        claims = []
+        has_unverified = False
+
+        # Simple sentence splitting (naive, but sufficient for MVP)
+        sentences = [s.strip() for s in re.split(r'[.!?]\s+', response_text) if s.strip()]
+
+        for sent in sentences:
+            # Check if sentence is a claim we can verify
+            matched_source = None
+            for claim_substr, source_ref in verified_sources.items():
+                if claim_substr.lower() in sent.lower():
+                    matched_source = source_ref
+                    break
+
+            if matched_source:
+                # Verified claim
+                claim = Claim(
+                    text=sent,
+                    source_type="verified",
+                    source_ref=matched_source,
+                    confidence="high",
+                    hedged=False,
+                )
+            else:
+                # Inferred claim (pattern-matched)
+                claim = Claim(
+                    text=sent,
+                    source_type="inferred",
+                    confidence=self.default_confidence,
+                    hedged=self._has_hedge(sent),
+                )
+                if not claim.hedged:
+                    has_unverified = True
+
+            claims.append(claim)
+
+        # Render the annotated response
+        rendered = self._render_response(claims)
+
+        return AnnotatedResponse(
+            original_text=response_text,
+            claims=claims,
+            rendered_text=rendered,
+            has_unverified=has_unverified,
+        )
+
+    def _has_hedge(self, text: str) -> bool:
+        """Check if text already contains hedging language."""
+        text_lower = text.lower()
+        for prefix in self.HEDGE_PREFIXES:
+            if text_lower.startswith(prefix.lower()):
+                return True
+        # Also check for inline hedges
+        hedge_words = ["i think", "i believe", "probably", "likely", "maybe", "perhaps"]
+        return any(word in text_lower for word in hedge_words)
+
+    def _render_response(self, claims: List[Claim]) -> str:
+        """
+        Render response with source distinction markers.
+
+        Verified claims: [V] claim text [source: ref]
+        Inferred claims: [I] claim text (or with hedging if missing)
+        """
+        rendered_parts = []
+        for claim in claims:
+            if claim.source_type == "verified":
+                part = f"[V] {claim.text}"
+                if claim.source_ref:
+                    part += f" [source: {claim.source_ref}]"
+            else:  # inferred
+                if not claim.hedged:
+                    # Add hedging if missing
+                    hedged_text = f"I think {claim.text[0].lower()}{claim.text[1:]}" if claim.text else claim.text
+                    part = f"[I] {hedged_text}"
+                else:
+                    part = f"[I] {claim.text}"
+            rendered_parts.append(part)
+        return " ".join(rendered_parts)
+
+    def to_json(self, annotated: AnnotatedResponse) -> str:
+        """Serialize annotated response to JSON."""
+        return json.dumps(
+            {
+                "original_text": annotated.original_text,
+                "rendered_text": annotated.rendered_text,
+                "has_unverified": annotated.has_unverified,
+                "claims": [asdict(c) for c in annotated.claims],
+            },
+            indent=2,
+            ensure_ascii=False,
+        )

tests/test_hardware_mcp_functional.py

@@ -1,51 +0,0 @@
-#!/usr/bin/env python3
-"""Functional test for hardware_mcp_server — uses asyncio.get_event_loop for restricted envs."""
-
-import asyncio, json, tempfile, sys
-from pathlib import Path
-sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
-from scripts.hardware_mcp_server import call_tool, is_path_allowed
-
-async def run_tests():
-    # Path allowlist
-    assert is_path_allowed(Path.home() / "any.txt")
-    assert is_path_allowed(Path("/tmp/foo"))
-    assert not is_path_allowed(Path("/etc/passwd"))
-    print("✓ Path allowlist")
-
-    # file_list on home
-    res = await call_tool("file_list", {"directory": "~"})
-    data = json.loads(res[0].text)
-    assert "entries" in data and data["count"] >= 0
-    print(f"✓ file_list works, entries: {data['count']}")
-
-    # file_write + file_read round-trip in temp dir
-    with tempfile.TemporaryDirectory() as td:
-        fp = Path(td) / "hmcp_test.txt"
-        content = "Hardware MCP round-trip OK"
-        w = await call_tool("file_write", {"path": str(fp), "content": content})
-        assert json.loads(w[0].text).get("success")
-        r = await call_tool("file_read", {"path": str(fp)})
-        assert json.loads(r[0].text)["content"] == content
-    print("✓ file write/read round-trip")
-
-    # file_read error: missing file
-    err = await call_tool("file_read", {"path": str(Path.home() / "no_such_file_xyz")})
-    assert "error" in json.loads(err[0].text)
-    print("✓ file_read reports missing file")
-
-    # Security: path traversal blocked
-    block = await call_tool("file_read", {"path": "/etc/passwd"})
-    bd = json.loads(block[0].text)
-    assert "not allowed" in bd.get("error", "").lower()
-    print("✓ Path traversal blocked")
-
-    print("\nAll functional checks passed!")
-
-if __name__ == "__main__":
-    # Use get_event_loop for environments where asyncio.run is disabled
-    try:
-        asyncio.run(run_tests())
-    except RuntimeError:
-        loop = asyncio.get_event_loop()
-        loop.run_until_complete(run_tests())

tests/test_hardware_mcp_server.py

@@ -1,126 +0,0 @@
-#!/usr/bin/env python3
-"""Smoke tests for hardware_mcp_server."""
-
-import json
-import os
-import subprocess
-import sys
-import tempfile
-from pathlib import Path
-from unittest import TestCase
-
-# Add repo root to path
-ROOT = Path(__file__).resolve().parent.parent
-sys.path.insert(0, str(ROOT))
-
-
-class TestHardwareMCPToolDefinitions(TestCase):
-    """Verify the MCP server is well-formed and tools have required schemas."""
-
-    def test_server_imports(self):
-        """Server module must import cleanly."""
-        import importlib.util
-        spec = importlib.util.spec_from_file_location(
-            "hardware_mcp_server",
-            ROOT / "scripts" / "hardware_mcp_server.py"
-        )
-        self.assertIsNotNone(spec)
-        mod = importlib.util.module_from_spec(spec)
-        spec.loader.exec_module(mod)
-        self.assertTrue(hasattr(mod, "app"))
-        self.assertTrue(hasattr(mod, "list_tools"))
-        self.assertTrue(hasattr(mod, "call_tool"))
-
-    def test_list_tools_returns_at_least_five_tools(self):
-        """list_tools() must return multiple tools covering file ops, lights, and system info."""
-        import asyncio
-        from scripts.hardware_mcp_server import list_tools
-        tools = asyncio.run(list_tools())
-        tool_names = [t.name for t in tools]
-        # Core capabilities
-        self.assertIn("file_read", tool_names)
-        self.assertIn("file_write", tool_names)
-        self.assertIn("file_list", tool_names)
-        self.assertIn("light_list", tool_names)
-        self.assertIn("light_control", tool_names)
-        self.assertIn("room_control", tool_names)
-        self.assertIn("scene_set", tool_names)
-        self.assertIn("system_info", tool_names)
-        self.assertGreaterEqual(len(tools), 8)
-
-    def test_file_read_schema_requires_path(self):
-        """file_read tool must require 'path' parameter."""
-        import asyncio
-        from scripts.hardware_mcp_server import list_tools
-        tools = asyncio.run(list_tools())
-        ft = next(t for t in tools if t.name == "file_read")
-        self.assertIn("path", ft.inputSchema["properties"])
-        self.assertIn("path", ft.inputSchema["required"])
-
-    def test_light_control_schema_requires_name_and_on(self):
-        """light_control requires name and on."""
-        import asyncio
-        from scripts.hardware_mcp_server import list_tools
-        tools = asyncio.run(list_tools())
-        ft = next(t for t in tools if t.name == "light_control")
-        self.assertIn("name", ft.inputSchema["required"])
-        self.assertIn("on", ft.inputSchema["required"])
-
-    def test_system_info_is_readonly(self):
-        """system_info tool takes no arguments."""
-        import asyncio
-        from scripts.hardware_mcp_server import list_tools
-        tools = asyncio.run(list_tools())
-        ft = next(t for t in tools if t.name == "system_info")
-        self.assertEqual(ft.inputSchema.get("required", []), [])
-        self.assertEqual(len(ft.inputSchema.get("properties", {})), 0)
-
-    def test_file_write_path_allowed_check(self):
-        """File write must enforce path allowlist (regression guard)."""
-        from scripts.hardware_mcp_server import is_path_allowed, Path
-        self.assertTrue(is_path_allowed(Path.home() / "test.txt"))
-        self.assertTrue(is_path_allowed(Path("/tmp/test.txt")))
-        # Outside allowed dirs should be rejected
-        self.assertFalse(is_path_allowed(Path("/etc/passwd")))
-
-    def test_run_openhue_error_handling(self):
-        """openhue runner returns structured error when CLI missing."""
-        from scripts.hardware_mcp_server import run_openhue
-        result = run_openhue(["get", "light"])
-        # On a system without openhue, must return success=False with helpful error
-        self.assertIn("success", result)
-        if not result.get("success"):
-            self.assertIn("error", result)
-            self.assertIn("openhue", result.get("error", "").lower())
-
-
-class TestHardwareMCPConfigCompleteness(TestCase):
-    """Validate config template matches tool set."""
-
-    def test_config_template_exists(self):
-        self.assertTrue((ROOT / "timmy-local" / "hardware_mcp_config.yaml").exists())
-
-    def test_config_lists_all_tools(self):
-        with open(ROOT / "timmy-local" / "hardware_mcp_config.yaml") as f:
-            content = f.read()
-        # All tool names should appear in the tools: section
-        for tool in ["file_read", "file_write", "file_list", "light_list",
-                     "light_control", "room_control", "scene_set", "system_info"]:
-            self.assertIn(tool, content, f"Tool {tool} missing from config tools list")
-
-    def test_config_has_security_guards(self):
-        with open(ROOT / "timmy-local" / "hardware_mcp_config.yaml") as f:
-            content = f.read()
-        self.assertIn("max_consecutive_errors", content)
-        self.assertIn("allowed_dirs", content)
-        self.assertIn("max_file_size_bytes", content)
-
-    def test_config_has_server_key(self):
-        with open(ROOT / "timmy-local" / "hardware_mcp_config.yaml") as f:
-            content = f.read()
-        self.assertIn("server_key: hardware", content)
-
-
-if __name__ == "__main__":
-    import unittest
-    unittest.main()

tests/timmy/test_claim_annotator.py

@@ -0,0 +1,103 @@
+#!/usr/bin/env python3
+"""Tests for claim_annotator.py — verifies source distinction is present."""
+
+import sys
+import os
+import json
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "src"))
+
+from timmy.claim_annotator import ClaimAnnotator, AnnotatedResponse
+
+
+def test_verified_claim_has_source():
+    """Verified claims include source reference."""
+    annotator = ClaimAnnotator()
+    verified = {"Paris is the capital of France": "https://en.wikipedia.org/wiki/Paris"}
+    response = "Paris is the capital of France. It is a beautiful city."
+
+    result = annotator.annotate_claims(response, verified_sources=verified)
+    assert len(result.claims) > 0
+    verified_claims = [c for c in result.claims if c.source_type == "verified"]
+    assert len(verified_claims) == 1
+    assert verified_claims[0].source_ref == "https://en.wikipedia.org/wiki/Paris"
+    assert "[V]" in result.rendered_text
+    assert "[source:" in result.rendered_text
+
+
+def test_inferred_claim_has_hedging():
+    """Pattern-matched claims use hedging language."""
+    annotator = ClaimAnnotator()
+    response = "The weather is nice today. It might rain tomorrow."
+
+    result = annotator.annotate_claims(response)
+    inferred_claims = [c for c in result.claims if c.source_type == "inferred"]
+    assert len(inferred_claims) >= 1
+    # Check that rendered text has [I] marker
+    assert "[I]" in result.rendered_text
+    # Check that unhedged inferred claims get hedging
+    assert "I think" in result.rendered_text or "I believe" in result.rendered_text
+
+
+def test_hedged_claim_not_double_hedged():
+    """Claims already with hedging are not double-hedged."""
+    annotator = ClaimAnnotator()
+    response = "I think the sky is blue. It is a nice day."
+
+    result = annotator.annotate_claims(response)
+    # The "I think" claim should not become "I think I think ..."
+    assert "I think I think" not in result.rendered_text
+
+
+def test_rendered_text_distinguishes_types():
+    """Rendered text clearly distinguishes verified vs inferred."""
+    annotator = ClaimAnnotator()
+    verified = {"Earth is round": "https://science.org/earth"}
+    response = "Earth is round. Stars are far away."
+
+    result = annotator.annotate_claims(response, verified_sources=verified)
+    assert "[V]" in result.rendered_text  # verified marker
+    assert "[I]" in result.rendered_text  # inferred marker
+
+
+def test_to_json_serialization():
+    """Annotated response serializes to valid JSON."""
+    annotator = ClaimAnnotator()
+    response = "Test claim."
+    result = annotator.annotate_claims(response)
+    json_str = annotator.to_json(result)
+    parsed = json.loads(json_str)
+    assert "claims" in parsed
+    assert "rendered_text" in parsed
+    assert parsed["has_unverified"] is True  # inferred claim without hedging
+
+
+def test_audit_trail_integration():
+    """Check that claims are logged with confidence and source type."""
+    # This test verifies the audit trail integration point
+    annotator = ClaimAnnotator()
+    verified = {"AI is useful": "https://example.com/ai"}
+    response = "AI is useful. It can help with tasks."
+
+    result = annotator.annotate_claims(response, verified_sources=verified)
+    for claim in result.claims:
+        assert claim.source_type in ("verified", "inferred")
+        assert claim.confidence in ("high", "medium", "low", "unknown")
+        if claim.source_type == "verified":
+            assert claim.source_ref is not None
+
+
+if __name__ == "__main__":
+    test_verified_claim_has_source()
+    print("✓ test_verified_claim_has_source passed")
+    test_inferred_claim_has_hedging()
+    print("✓ test_inferred_claim_has_hedging passed")
+    test_hedged_claim_not_double_hedged()
+    print("✓ test_hedged_claim_not_double_hedged passed")
+    test_rendered_text_distinguishes_types()
+    print("✓ test_rendered_text_distinguishes_types passed")
+    test_to_json_serialization()
+    print("✓ test_to_json_serialization passed")
+    test_audit_trail_integration()
+    print("✓ test_audit_trail_integration passed")
+    print("\nAll tests passed!")

timmy-local/hardware/README.md

@@ -1,3 +0,0 @@
-# hardware MCP config
-
-Copy `hardware_mcp_config.yaml` to `~/.timmy/hardware/hardware_mcp_config.yaml` to enable runtime tuning.

timmy-local/hardware_mcp_config.yaml

@@ -1,67 +0,0 @@
-# ═══════════════════════════════════════════════════════════════════════
-# Local Hardware MCP — Runtime Configuration
-# ═══════════════════════════════════════════════════════════════════════
-# Edit this file to tune hardware control settings.
-# Hermes loads this at session start when the hardware MCP server is enabled.
-#
-# Location: ~/.timmy/hardware/hardware_mcp_config.yaml
-# ═══════════════════════════════════════════════════════════════════════
-
-# ── Server Identity ───────────────────────────────────────────────────
-server_key: hardware
-
-# ── Tool Names ────────────────────────────────────────────────────────
-# Exact tool names Hermes registers.  Update if you rename tools in
-# hardware_mcp_server.py.
-tools:
-  - name: file_read
-    hint: "Read a file from an allowed directory (home, /tmp). Max 10 MB."
-  - name: file_write
-    hint: "Write text content to a file within allowed directories."
-  - name: file_list
-    hint: "List files and directories in a given folder."
-  - name: light_list
-    hint: "List all Hue lights, rooms, and scenes from OpenHue."
-  - name: light_control
-    hint: "Control a specific Hue light: on/off, brightness, color, temperature."
-  - name: room_control
-    hint: "Control all lights in a room: on/off, brightness."
-  - name: scene_set
-    hint: "Activate a Hue scene in a room."
-  - name: system_info
-    hint: "Get safe system information: OS, CPU count, memory usage, disk space."
-
-# ── Security Guards ───────────────────────────────────────────────────
-guards:
-  # Maximum consecutive tool errors before stopping.
-  max_consecutive_errors: 3
-
-  # Max total hardware MCP calls per session (0 = unlimited).
-  max_mcp_calls_per_session: 0
-
-  # Allowed directories for file operations (expanded paths).
-  allowed_dirs:
-    - "~"
-    - "/tmp"
-    - "/private/tmp"
-
-  # Maximum file size for reads (bytes).
-  max_file_size_bytes: 10485760  # 10 MB
-
-# ── OpenHue ───────────────────────────────────────────────────────────
-# Path to openhue CLI (auto-detected if in PATH).
-openhue_command: "openhue"
-
-# ── Dependencies ───────────────────────────────────────────────────────
-# Prerequisites:
-#   - OpenHue CLI: brew install openhue/cli/openhue (macOS) or see https://github.com/openhue/openhue-cli
-#   - MCP SDK: pip install mcp
-#   - For system_info: pip install psutil (optional, for detailed memory/disk metrics)
-#
-# Config in ~/.hermes/config.yaml:
-#   mcp_servers:
-#     hardware:
-#       command: "python"
-#       args: ["/Users/you/path/to/timmy-home/scripts/hardware_mcp_server.py"]
-#       env:
-#         OPENHUE_BRIDGE_IP: "192.168.1.xx"  # optional, if openhue needs it