commit 3a62428751cf5b34dd86e3c61c8173bd263db350
Author: brian <brian@truenas.local>
Date:   Fri May 29 14:38:57 2026 -0700

    Initial commit: Handoff Pro MCP server for Kellow Construction

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..2fc63c9
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,4 @@
+__pycache__/
+*.pyc
+data/*.db
+.env
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..63e7619
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,8 @@
+FROM python:3.12-slim
+WORKDIR /app
+COPY lib/ lib/
+COPY scripts/ scripts/
+COPY server.py .
+RUN python3 scripts/init_db.py
+EXPOSE 3101
+CMD ["python3", "server.py"]
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..7397a32
--- /dev/null
+++ b/README.md
@@ -0,0 +1,53 @@
+# Handoff Pro — Construction Project Management MCP Server
+
+MCP server extending Tom (Kellow Construction AI agent on GW-2) with estimation, proposals, invoicing, and project management — synced with JobTread + QuickBooks Online.
+
+## Architecture
+
+```
+Tom (Telegram) → mcporter → handoff-pro MCP (port 3101)
+                          → design-agent MCP (port 3100)
+                          ↓
+                    SQLite (local) ←→ JobTread API
+                                   ←→ QuickBooks Online API
+```
+
+## Quick Start
+
+```bash
+# Run tests
+cd handoff-pro && python3 -m pytest tests/
+
+# Run server locally
+python3 server.py
+
+# Deploy to GW-2
+chmod +x deploy.sh && ./deploy.sh
+
+# Or run as Docker container
+docker build -t handoff-pro .
+docker run -d --name handoff-pro -p 3101:3101 \
+  -v /path/to/.jobtread-api.json:/app/data/.jobtread-api.json:ro \
+  handoff-pro
+```
+
+## Tools (34 total)
+
+See `SKILL.md` for full tool reference and workflow patterns.
+
+## Dependencies
+
+- Python 3.10+ (stdlib only — no pip install needed)
+- SQLite3 (bundled with Python)
+- Network access to JobTread API and QBO token server
+
+## Deployment Options
+
+1. **In-container** (recommended): Deploy scripts directly into Tom's workspace via `deploy.sh`
+2. **Standalone Docker**: Run as separate container on TrueNAS, register in mcporter
+3. **Hybrid**: Scripts in workspace + server as sidecar container
+
+## Credentials
+
+- **JobTread**: `.jobtread-api.json` with `grantKey` and `organizationId`
+- **QBO**: Token server at `http://192.168.86.11:18801/kellow-tokens.json`
diff --git a/SKILL.md b/SKILL.md
new file mode 100644
index 0000000..b95461e
--- /dev/null
+++ b/SKILL.md
@@ -0,0 +1,107 @@
+---
+name: handoff-pro
+description: "Construction estimation, proposals, invoicing, and project management for Kellow Construction. Syncs with JobTread and QuickBooks Online. Use for any estimate, proposal, invoice, budget, or project tracking request."
+---
+
+# Handoff Pro — Construction Project Management
+
+MCP server at `http://192.168.86.11:3101/sse` (registered in mcporter as `handoff-pro`).
+
+## Quick Reference
+
+Call tools via: `mcporter call handoff-pro.<tool> key=value`
+
+### Estimation Workflow
+1. User describes job → `generate_estimate description="kitchen remodel, 200sqft, new cabinets and countertops"`
+2. Optionally enrich with design-agent: `mcporter call design-agent.estimate_cost ...` and `design-agent.get_permit_matrix ...`
+3. Review estimate → `get_estimate estimate_id=<id>`
+4. Generate proposal → `draft_proposal estimate_id=<id>`
+5. Request signature → `request_signature estimate_id=<id> client_name="John Smith"`
+
+### Invoice Workflow
+1. Create invoice → `create_invoice estimate_id=<id>`
+2. Sync to JobTread → `jt_push_invoice invoice_id=<id>`
+3. Sync to QBO → `sync_invoice_qbo invoice_id=<id>`
+4. Mark paid → `mark_invoice_paid invoice_id=<id>`
+
+### JobTread Sync
+- Pull all jobs: `jt_pull_jobs`
+- List customers: `jt_list_customers`
+- Push estimate: `jt_push_estimate estimate_id=<id>`
+- Check sync: `jt_sync_status job_id=<id>`
+
+### Daily Operations
+- Log work: `create_daily_log job_id=<id> notes="framed walls" crew_names="Mike, Sarah" hours_worked=8`
+- Analyze site: `analyze_site observations="water damage in corner, cracked foundation" job_id=<id>`
+- Budget check: `get_budget_report job_id=<id>`
+- Set milestone: `set_milestone job_id=<id> phase_name="Framing" start_date="2026-06-01"`
+
+### Pricing Catalogs
+- List: `list_catalogs`
+- Create: `create_catalog name="Commercial" labor_rates={"framing":95,"electrical":110}`
+
+## All Tools (30)
+
+| Tool | Purpose |
+|------|---------|
+| `list_catalogs` | List pricing catalogs |
+| `create_catalog` | Create pricing catalog |
+| `get_catalog` | Get catalog by ID |
+| `update_catalog` | Update catalog rates |
+| `delete_catalog` | Delete catalog |
+| `create_job` | Create local job |
+| `list_jobs` | List jobs (optional status filter) |
+| `get_job` | Get job + estimates + invoices |
+| `update_job` | Update job fields |
+| `create_estimate` | Create estimate with line items |
+| `get_estimate` | Get estimate + line items |
+| `generate_estimate` | AI: parse description → estimate |
+| `draft_proposal` | Generate proposal from estimate |
+| `create_invoice` | Create invoice from estimate |
+| `list_invoices` | List invoices |
+| `mark_invoice_paid` | Mark invoice paid |
+| `sync_invoice_qbo` | Sync invoice to QuickBooks |
+| `jt_list_customers` | List JobTread customers |
+| `jt_pull_jobs` | Pull jobs from JobTread |
+| `jt_push_estimate` | Push estimate to JobTread |
+| `jt_push_invoice` | Push invoice to JobTread |
+| `jt_sync_status` | Get sync status |
+| `generate_scope` | Generate scope of work |
+| `generate_material_list` | Generate shopping list |
+| `create_daily_log` | Log daily work |
+| `list_daily_logs` | List logs for job |
+| `analyze_site` | Analyze site conditions |
+| `get_budget_report` | Budget vs actual report |
+| `set_milestone` | Set project milestone |
+| `get_schedule` | Get project timeline |
+| `request_signature` | Request e-signature |
+| `record_signature` | Record signature |
+| `log_activity` | Log client action |
+| `get_activity` | Get activity history |
+
+## Design-Agent Integration
+
+For SB-area jobs, enrich estimates with:
+- `mcporter call design-agent.estimate_cost` — local cost ranges
+- `mcporter call design-agent.get_permit_matrix` — permit fees + timeline
+- `mcporter call design-agent.lookup_parcel` — lot data affecting scope
+
+## Conversation Patterns
+
+**"I need an estimate for X"** → Ask for address + details, then `generate_estimate`. If SB area, also call design-agent.estimate_cost for validation.
+
+**"Send a proposal"** → `draft_proposal` → format nicely → send to client. Ask if they want signature collection.
+
+**"Invoice this job"** → `create_invoice` → ask about syncing to JobTread/QBO.
+
+**"What's the budget looking like?"** → `get_budget_report` → present variance and flags.
+
+**"Log today's work"** → Ask for crew, hours, notes, photos → `create_daily_log`.
+
+**"Show me my jobs"** → `list_jobs` or `jt_pull_jobs` for fresh data from JobTread.
+
+## Error Handling
+
+- If JobTread sync fails: tool returns error, work continues locally. Retry later.
+- If QBO sync fails: check token with `python3 skills/kellow-qbo/scripts/check-token.py`
+- All data persists in local SQLite regardless of sync status.
diff --git a/deploy.sh b/deploy.sh
new file mode 100644
index 0000000..7cb58d6
--- /dev/null
+++ b/deploy.sh
@@ -0,0 +1,66 @@
+#!/bin/bash
+# Deploy Handoff Pro to Tom's workspace on GW-2
+# Usage: ./deploy.sh [--dry-run]
+set -e
+
+CONTAINER="openclaw-repo-openclaw-gateway-2-1"
+TARGET="/home/node/.openclaw/workspace/agents/tom/skills/handoff-pro"
+SOURCE="$(cd "$(dirname "$0")" && pwd)"
+DRY_RUN=""
+
+if [ "$1" = "--dry-run" ]; then DRY_RUN="echo [DRY-RUN]"; fi
+
+echo "=== Handoff Pro Deployment ==="
+echo "Source: $SOURCE"
+echo "Target: $CONTAINER:$TARGET"
+echo ""
+
+# Create target directory
+$DRY_RUN docker exec $CONTAINER mkdir -p $TARGET/{scripts,lib,templates,data,tests}
+
+# Copy core files
+for f in SKILL.md server.py; do
+    $DRY_RUN docker cp "$SOURCE/$f" "$CONTAINER:$TARGET/$f"
+done
+
+# Copy lib
+for f in "$SOURCE"/lib/*.py; do
+    $DRY_RUN docker cp "$f" "$CONTAINER:$TARGET/lib/$(basename $f)"
+done
+
+# Copy scripts
+for f in "$SOURCE"/scripts/*.py; do
+    $DRY_RUN docker cp "$f" "$CONTAINER:$TARGET/scripts/$(basename $f)"
+done
+
+# Copy templates
+for f in "$SOURCE"/templates/*; do
+    [ -f "$f" ] && $DRY_RUN docker cp "$f" "$CONTAINER:$TARGET/templates/$(basename $f)"
+done
+
+# Fix permissions (no world-writable)
+$DRY_RUN docker exec $CONTAINER find $TARGET -type f -exec chmod 644 {} \;
+$DRY_RUN docker exec $CONTAINER find $TARGET -type d -exec chmod 755 {} \;
+$DRY_RUN docker exec $CONTAINER chown -R node:node $TARGET
+
+# Initialize database
+$DRY_RUN docker exec -u node $CONTAINER python3 $TARGET/scripts/init_db.py
+
+# Add to mcporter config
+$DRY_RUN docker exec -u node $CONTAINER sh -c "
+    CONFIG=/home/node/.openclaw/workspace/config/mcporter.json
+    python3 -c \"
+import json
+with open('\$CONFIG') as f: cfg = json.load(f)
+cfg['mcpServers']['handoff-pro'] = {'url': 'http://127.0.0.1:3101/sse'}
+with open('\$CONFIG','w') as f: json.dump(cfg, f, indent=4)
+print('mcporter.json updated')
+\"
+"
+
+echo ""
+echo "=== Deployment complete ==="
+echo "Next steps:"
+echo "  1. Start the MCP server: docker exec -u node -d $CONTAINER python3 $TARGET/server.py"
+echo "  2. Verify: docker exec $CONTAINER curl -s http://127.0.0.1:3101/health"
+echo "  3. Test: mcporter call handoff-pro.list_catalogs"
diff --git a/lib/__init__.py b/lib/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/lib/db.py b/lib/db.py
new file mode 100644
index 0000000..9f7db40
--- /dev/null
+++ b/lib/db.py
@@ -0,0 +1,156 @@
+"""SQLite database initialization and access for Handoff Pro."""
+import json
+import os
+import sqlite3
+import uuid
+from datetime import datetime
+
+DB_PATH = os.environ.get("HANDOFF_DB",
+    os.path.join(os.path.dirname(__file__), "..", "data", "handoff.db"))
+
+SCHEMA = """
+CREATE TABLE IF NOT EXISTS jobs (
+  id TEXT PRIMARY KEY,
+  user_id TEXT DEFAULT 'kellow',
+  jobtread_job_id TEXT UNIQUE,
+  client_name TEXT,
+  jobtread_customer_id TEXT,
+  address TEXT,
+  description TEXT,
+  status TEXT DEFAULT 'proposal',
+  created_at TEXT,
+  updated_at TEXT,
+  synced_to_jobtread_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS estimates (
+  id TEXT PRIMARY KEY,
+  job_id TEXT REFERENCES jobs(id),
+  jobtread_estimate_id TEXT UNIQUE,
+  labor_hours REAL DEFAULT 0,
+  labor_rate REAL DEFAULT 0,
+  materials_cost REAL DEFAULT 0,
+  markup_percent REAL DEFAULT 20,
+  total_cost REAL DEFAULT 0,
+  notes TEXT,
+  status TEXT DEFAULT 'draft',
+  created_at TEXT,
+  synced_to_jobtread_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS line_items (
+  id TEXT PRIMARY KEY,
+  estimate_id TEXT REFERENCES estimates(id),
+  description TEXT,
+  category TEXT,
+  quantity REAL DEFAULT 1,
+  unit TEXT DEFAULT 'ea',
+  unit_cost REAL DEFAULT 0,
+  total REAL DEFAULT 0
+);
+
+CREATE TABLE IF NOT EXISTS pricing_catalogs (
+  id TEXT PRIMARY KEY,
+  user_id TEXT DEFAULT 'kellow',
+  name TEXT,
+  labor_rates TEXT,
+  material_markups TEXT,
+  created_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS invoices (
+  id TEXT PRIMARY KEY,
+  estimate_id TEXT REFERENCES estimates(id),
+  job_id TEXT REFERENCES jobs(id),
+  invoice_number TEXT,
+  jobtread_invoice_id TEXT UNIQUE,
+  qbo_invoice_id TEXT,
+  amount_due REAL DEFAULT 0,
+  amount_paid REAL DEFAULT 0,
+  due_date TEXT,
+  status TEXT DEFAULT 'draft',
+  created_at TEXT,
+  synced_to_jobtread_at TEXT,
+  synced_to_qbo_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS daily_logs (
+  id TEXT PRIMARY KEY,
+  job_id TEXT REFERENCES jobs(id),
+  log_date TEXT,
+  crew_names TEXT,
+  hours_worked REAL DEFAULT 0,
+  notes TEXT,
+  photos TEXT,
+  created_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS signatures (
+  id TEXT PRIMARY KEY,
+  estimate_id TEXT REFERENCES estimates(id),
+  client_name TEXT,
+  signature_data TEXT,
+  signed_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS client_activity (
+  id TEXT PRIMARY KEY,
+  estimate_id TEXT REFERENCES estimates(id),
+  job_id TEXT REFERENCES jobs(id),
+  action TEXT,
+  metadata TEXT,
+  timestamp TEXT
+);
+
+CREATE TABLE IF NOT EXISTS jobtread_sync_log (
+  id TEXT PRIMARY KEY,
+  local_id TEXT,
+  jobtread_id TEXT,
+  entity_type TEXT,
+  action TEXT,
+  status TEXT DEFAULT 'pending',
+  error_message TEXT,
+  synced_at TEXT
+);
+
+CREATE TABLE IF NOT EXISTS project_milestones (
+  id TEXT PRIMARY KEY,
+  job_id TEXT REFERENCES jobs(id),
+  phase_name TEXT,
+  start_date TEXT,
+  end_date TEXT,
+  status TEXT DEFAULT 'pending'
+);
+"""
+
+def get_db() -> sqlite3.Connection:
+    os.makedirs(os.path.dirname(DB_PATH), exist_ok=True)
+    conn = sqlite3.connect(DB_PATH, timeout=10)
+    conn.row_factory = sqlite3.Row
+    conn.execute("PRAGMA journal_mode=WAL")
+    conn.execute("PRAGMA busy_timeout=5000")
+    return conn
+
+def init_db():
+    conn = get_db()
+    conn.executescript(SCHEMA)
+    conn.commit()
+    conn.close()
+
+def new_id():
+    return str(uuid.uuid4())[:8]
+
+def now():
+    return datetime.now().astimezone().isoformat()
+
+def row_to_dict(row):
+    if row is None:
+        return None
+    return dict(row)
+
+def rows_to_list(rows):
+    return [dict(r) for r in rows]
+
+if __name__ == "__main__":
+    init_db()
+    print(json.dumps({"ok": True, "db_path": DB_PATH}))
diff --git a/lib/pave.py b/lib/pave.py
new file mode 100644
index 0000000..d2d165c
--- /dev/null
+++ b/lib/pave.py
@@ -0,0 +1,41 @@
+"""JobTread Pave API helper. Reads creds from .jobtread-api.json."""
+import json
+import os
+import urllib.request
+
+PAVE_URL = "https://api.jobtread.com/pave"
+_creds = None
+
+def _get_creds():
+    global _creds
+    if _creds:
+        return _creds
+    paths = [
+        os.environ.get("JOBTREAD_CREDS"),
+        os.path.join(os.path.dirname(__file__), "..", "data", ".jobtread-api.json"),
+        "/home/node/.openclaw/workspace/agents/tom/.jobtread-api.json",
+    ]
+    for p in paths:
+        if p and os.path.exists(p):
+            with open(p) as f:
+                _creds = json.load(f)
+            return _creds
+    raise RuntimeError("No .jobtread-api.json found")
+
+def get_grant_key():
+    return _get_creds()["grantKey"]
+
+def get_org_id():
+    return _get_creds()["organizationId"]
+
+def pave_query(query: dict) -> dict:
+    """Execute a Pave API query. Injects grantKey automatically."""
+    payload = {"query": {"$": {"grantKey": get_grant_key()}, **query}}
+    data = json.dumps(payload).encode()
+    req = urllib.request.Request(PAVE_URL, data=data,
+        headers={"Content-Type": "application/json"}, method="POST")
+    with urllib.request.urlopen(req, timeout=30) as resp:
+        result = json.loads(resp.read())
+    if "errors" in result:
+        raise RuntimeError(f"Pave API error: {result['errors']}")
+    return result
diff --git a/lib/qbo.py b/lib/qbo.py
new file mode 100644
index 0000000..1eb3e86
--- /dev/null
+++ b/lib/qbo.py
@@ -0,0 +1,54 @@
+"""QuickBooks Online API helper. Tokens from http://192.168.86.11:18801."""
+import json
+import time
+import urllib.request
+
+TOKEN_URL = "http://192.168.86.11:18801/kellow-tokens.json"
+REFRESH_URL = "http://192.168.86.11:18803/qbo/refresh?tenant=kellow"
+QBO_BASE = "https://quickbooks.api.intuit.com/v3/company"
+
+def get_token():
+    """Fetch valid QBO access token + realm_id, auto-refresh if expired."""
+    with urllib.request.urlopen(TOKEN_URL, timeout=10) as resp:
+        tokens = json.loads(resp.read())
+    if time.time() >= tokens.get("expires_at_epoch", 0):
+        req = urllib.request.Request(REFRESH_URL, method="POST")
+        urllib.request.urlopen(req, timeout=15)
+        with urllib.request.urlopen(TOKEN_URL, timeout=10) as resp:
+            tokens = json.loads(resp.read())
+    return tokens["access_token"], tokens["realm_id"]
+
+def qbo_request(method, endpoint, body=None):
+    """Make authenticated QBO API request. Returns parsed JSON."""
+    access_token, realm_id = get_token()
+    url = f"{QBO_BASE}/{realm_id}/{endpoint}"
+    data = json.dumps(body).encode() if body else None
+    req = urllib.request.Request(url, data=data, method=method, headers={
+        "Authorization": f"Bearer {access_token}",
+        "Accept": "application/json",
+        "Content-Type": "application/json",
+    })
+    with urllib.request.urlopen(req, timeout=30) as resp:
+        return json.loads(resp.read())
+
+def create_invoice(customer_id, line_items, due_date=None):
+    """Create invoice in QBO. line_items: [{description, amount, qty}]"""
+    lines = []
+    for i, item in enumerate(line_items, 1):
+        lines.append({
+            "LineNum": i,
+            "Amount": item["amount"],
+            "DetailType": "SalesItemLineDetail",
+            "Description": item.get("description", ""),
+            "SalesItemLineDetail": {
+                "Qty": item.get("qty", 1),
+                "UnitPrice": item["amount"] / item.get("qty", 1),
+            }
+        })
+    invoice = {
+        "CustomerRef": {"value": customer_id},
+        "Line": lines,
+    }
+    if due_date:
+        invoice["DueDate"] = due_date
+    return qbo_request("POST", "invoice", invoice)
diff --git a/scripts/__init__.py b/scripts/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/scripts/activity_logger.py b/scripts/activity_logger.py
new file mode 100644
index 0000000..df4a1da
--- /dev/null
+++ b/scripts/activity_logger.py
@@ -0,0 +1,30 @@
+"""Client activity logger: track views, approvals, signatures, payments."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, rows_to_list
+
+def log(action, estimate_id=None, job_id=None, metadata=None):
+    aid = new_id()
+    conn = get_db()
+    conn.execute("INSERT INTO client_activity (id, estimate_id, job_id, action, metadata, timestamp) VALUES (?,?,?,?,?,?)",
+        (aid, estimate_id, job_id, action, json.dumps(metadata) if metadata else None, now()))
+    conn.commit(); conn.close()
+    return {"id": aid, "action": action, "timestamp": now()}
+
+def get(estimate_id=None, job_id=None):
+    conn = get_db()
+    if estimate_id:
+        rows = conn.execute("SELECT * FROM client_activity WHERE estimate_id=? ORDER BY timestamp DESC", (estimate_id,)).fetchall()
+    elif job_id:
+        rows = conn.execute("SELECT * FROM client_activity WHERE job_id=? ORDER BY timestamp DESC", (job_id,)).fetchall()
+    else:
+        rows = conn.execute("SELECT * FROM client_activity ORDER BY timestamp DESC LIMIT 50").fetchall()
+    conn.close()
+    return {"activity": rows_to_list(rows)}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "get"
+    if cmd == "log": print(json.dumps(log(sys.argv[2], estimate_id=sys.argv[3] if len(sys.argv)>3 else None)))
+    elif cmd == "get": print(json.dumps(get(estimate_id=sys.argv[2] if len(sys.argv)>2 else None), indent=2))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/budget_tracker.py b/scripts/budget_tracker.py
new file mode 100644
index 0000000..7dd2f9e
--- /dev/null
+++ b/scripts/budget_tracker.py
@@ -0,0 +1,40 @@
+"""Budget tracker: compare estimated vs actual costs."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, row_to_dict, rows_to_list
+
+def report(job_id):
+    conn = get_db()
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (job_id,)).fetchone())
+    if not job: conn.close(); return {"error": "job not found"}
+    estimates = rows_to_list(conn.execute("SELECT * FROM estimates WHERE job_id=?", (job_id,)).fetchall())
+    logs = rows_to_list(conn.execute("SELECT * FROM daily_logs WHERE job_id=?", (job_id,)).fetchall())
+    invoices = rows_to_list(conn.execute("SELECT * FROM invoices WHERE job_id=?", (job_id,)).fetchall())
+    conn.close()
+
+    budgeted = sum(e["total_cost"] for e in estimates) if estimates else 0
+    actual_hours = sum(l["hours_worked"] for l in logs)
+    budgeted_hours = sum(e["labor_hours"] for e in estimates) if estimates else 0
+    invoiced = sum(i["amount_due"] for i in invoices)
+    paid = sum(i["amount_paid"] or 0 for i in invoices)
+
+    hours_variance = actual_hours - budgeted_hours
+    hours_pct = (hours_variance / budgeted_hours * 100) if budgeted_hours else 0
+
+    flags = []
+    if hours_pct > 10: flags.append(f"⚠️ Labor {hours_pct:.0f}% over budget")
+    if paid < invoiced * 0.5 and invoices: flags.append("⚠️ Less than 50% collected")
+
+    return {
+        "job_id": job_id, "client": job.get("client_name"),
+        "budget": {"total_estimated": budgeted, "budgeted_hours": budgeted_hours},
+        "actual": {"hours_worked": actual_hours, "log_entries": len(logs)},
+        "financial": {"invoiced": invoiced, "collected": paid, "outstanding": invoiced - paid},
+        "variance": {"hours": hours_variance, "hours_pct": round(hours_pct, 1)},
+        "flags": flags
+    }
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    if len(sys.argv) < 2: print(json.dumps({"error": "usage: budget_tracker.py <job_id>"}))
+    else: print(json.dumps(report(sys.argv[1]), indent=2))
diff --git a/scripts/daily_log.py b/scripts/daily_log.py
new file mode 100644
index 0000000..c0e9b11
--- /dev/null
+++ b/scripts/daily_log.py
@@ -0,0 +1,43 @@
+"""Daily job log management."""
+import json, sys, os
+from datetime import date
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+
+def create(job_id, notes, crew_names=None, hours_worked=None, photos=None):
+    lid = new_id()
+    conn = get_db()
+    conn.execute("""INSERT INTO daily_logs (id, job_id, log_date, crew_names, hours_worked, notes, photos, created_at)
+        VALUES (?,?,?,?,?,?,?,?)""",
+        (lid, job_id, date.today().isoformat(), crew_names, hours_worked or 0, notes,
+         json.dumps(photos) if photos else None, now()))
+    conn.commit(); conn.close()
+    return {"id": lid, "job_id": job_id, "date": date.today().isoformat()}
+
+def list_logs(job_id):
+    conn = get_db()
+    rows = conn.execute("SELECT * FROM daily_logs WHERE job_id=? ORDER BY log_date DESC", (job_id,)).fetchall()
+    conn.close()
+    result = rows_to_list(rows)
+    for r in result:
+        r["photos"] = json.loads(r["photos"]) if r["photos"] else []
+    return {"logs": result}
+
+def get_log(log_id):
+    conn = get_db()
+    row = row_to_dict(conn.execute("SELECT * FROM daily_logs WHERE id=?", (log_id,)).fetchone())
+    conn.close()
+    if not row: return {"error": "log not found"}
+    row["photos"] = json.loads(row["photos"]) if row["photos"] else []
+    return row
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "list"
+    if cmd == "create":
+        print(json.dumps(create(sys.argv[2], sys.argv[3],
+            sys.argv[4] if len(sys.argv)>4 else None,
+            float(sys.argv[5]) if len(sys.argv)>5 else None)))
+    elif cmd == "list": print(json.dumps(list_logs(sys.argv[2]), indent=2))
+    elif cmd == "get": print(json.dumps(get_log(sys.argv[2]), indent=2))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/estimate_generator.py b/scripts/estimate_generator.py
new file mode 100644
index 0000000..625dab5
--- /dev/null
+++ b/scripts/estimate_generator.py
@@ -0,0 +1,108 @@
+"""Estimate generator: parse job description into structured estimate."""
+import json, sys, os, re
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, init_db
+from scripts.pricing_catalog import get_catalog, list_catalogs
+from scripts.project_mgmt import create_job, create_estimate
+
+# Common construction scope patterns → (category, unit, base_hours_per_unit)
+SCOPE_PATTERNS = {
+    r"cabinet": ("finish_carpentry", "lf", 2.5),
+    r"countertop": ("tile", "sf", 0.5),
+    r"floor": ("flooring", "sf", 0.15),
+    r"tile": ("tile", "sf", 0.25),
+    r"paint": ("painting", "sf", 0.05),
+    r"drywall": ("drywall", "sf", 0.1),
+    r"electric": ("electrical", "ea", 4),
+    r"plumb": ("plumbing", "ea", 6),
+    r"frame|wall": ("framing", "lf", 1.5),
+    r"demo": ("demolition", "sf", 0.08),
+    r"roof": ("roofing", "sf", 0.12),
+    r"concrete|foundation": ("concrete", "cy", 8),
+    r"window": ("finish_carpentry", "ea", 3),
+    r"door": ("finish_carpentry", "ea", 4),
+    r"deck": ("framing", "sf", 0.3),
+    r"bathroom": ("plumbing", "ea", 40),
+    r"kitchen": ("general_labor", "ea", 120),
+}
+
+# Material cost estimates per unit
+MATERIAL_COSTS = {
+    "cabinet": 250, "countertop": 75, "flooring": 8, "tile": 12,
+    "paint": 0.50, "drywall": 3, "electrical": 150, "plumbing": 200,
+    "framing": 5, "demolition": 2, "roofing": 6, "concrete": 180,
+    "window": 600, "door": 400, "deck": 15, "bathroom": 3000, "kitchen": 8000,
+}
+
+def generate(description, job_id=None, catalog_id=None, markup_percent=20):
+    """Generate estimate from text description."""
+    # Get pricing catalog
+    if catalog_id:
+        catalog = get_catalog(catalog_id)
+    else:
+        cats = list_catalogs()
+        catalog = cats["catalogs"][0] if cats["catalogs"] else None
+    rates = catalog.get("labor_rates", {}) if catalog else {}
+    markups = catalog.get("material_markups", {}) if catalog else {}
+
+    # Extract quantities from description
+    sqft_match = re.search(r"(\d+)\s*(?:sq\s*ft|sqft|sf)", description, re.I)
+    sqft = int(sqft_match.group(1)) if sqft_match else 200  # default
+
+    # Build line items from description
+    line_items = []
+    desc_lower = description.lower()
+    matched = False
+
+    for pattern, (category, unit, hours_per) in SCOPE_PATTERNS.items():
+        if re.search(pattern, desc_lower):
+            matched = True
+            qty = sqft if unit == "sf" else (sqft / 10 if unit == "lf" else 1)
+            labor_rate = rates.get(category, 75)
+            labor_hours = qty * hours_per
+            labor_cost = labor_hours * labor_rate
+
+            # Material cost
+            mat_key = next((k for k in MATERIAL_COSTS if re.search(pattern, k)), None)
+            mat_cost = MATERIAL_COSTS.get(mat_key, 10) * qty if mat_key else qty * 10
+            mat_markup = markups.get(category, 1.15)
+            mat_cost *= mat_markup
+
+            line_items.append({
+                "description": f"{category.replace('_',' ').title()} — {pattern.replace('|','/')}",
+                "category": "labor", "quantity": round(labor_hours, 1),
+                "unit": "hr", "unit_cost": labor_rate
+            })
+            line_items.append({
+                "description": f"Materials — {pattern.replace('|','/')}",
+                "category": "materials", "quantity": round(qty, 1),
+                "unit": unit, "unit_cost": round(mat_cost / qty, 2)
+            })
+
+    if not matched:
+        # Generic estimate based on sqft
+        line_items = [
+            {"description": "General Labor", "category": "labor", "quantity": sqft * 0.2, "unit": "hr", "unit_cost": rates.get("general_labor", 55)},
+            {"description": "Materials (general)", "category": "materials", "quantity": sqft, "unit": "sf", "unit_cost": 15},
+        ]
+
+    # Create job if needed
+    if not job_id:
+        job_result = create_job("TBD", description=description)
+        job_id = job_result["id"]
+
+    # Create estimate
+    result = create_estimate(job_id, line_items, markup_percent=markup_percent)
+    result["line_items_count"] = len(line_items)
+    result["description"] = description
+    return result
+
+if __name__ == "__main__":
+    init_db()
+    if len(sys.argv) < 2:
+        print(json.dumps({"error": "usage: estimate_generator.py <description> [catalog_id] [job_id]"}))
+    else:
+        desc = sys.argv[1]
+        cat_id = sys.argv[2] if len(sys.argv) > 2 else None
+        jid = sys.argv[3] if len(sys.argv) > 3 else None
+        print(json.dumps(generate(desc, job_id=jid, catalog_id=cat_id), indent=2))
diff --git a/scripts/init_db.py b/scripts/init_db.py
new file mode 100644
index 0000000..1e4cef6
--- /dev/null
+++ b/scripts/init_db.py
@@ -0,0 +1,11 @@
+#!/usr/bin/env python3
+"""Initialize Handoff Pro database and seed default data."""
+import sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import init_db
+from scripts.pricing_catalog import seed_default
+
+if __name__ == "__main__":
+    init_db()
+    result = seed_default()
+    print(f"Database initialized. Default catalog: {result}")
diff --git a/scripts/invoice_mgmt.py b/scripts/invoice_mgmt.py
new file mode 100644
index 0000000..3ee2a75
--- /dev/null
+++ b/scripts/invoice_mgmt.py
@@ -0,0 +1,81 @@
+"""Invoice management: create, track, sync to JobTread + QBO."""
+import json, sys, os
+from datetime import datetime, timedelta
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+
+def create(estimate_id, due_date=None):
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    # Generate invoice number
+    count = conn.execute("SELECT COUNT(*) as c FROM invoices").fetchone()["c"]
+    inv_num = f"KC-{count + 1001}"
+    if not due_date:
+        due_date = (datetime.now() + timedelta(days=30)).strftime("%Y-%m-%d")
+    inv_id = new_id()
+    conn.execute("""INSERT INTO invoices (id, estimate_id, job_id, invoice_number, amount_due, due_date, status, created_at)
+        VALUES (?,?,?,?,?,?,?,?)""",
+        (inv_id, estimate_id, est["job_id"], inv_num, est["total_cost"], due_date, "draft", now()))
+    conn.commit(); conn.close()
+    return {"id": inv_id, "invoice_number": inv_num, "amount_due": est["total_cost"], "due_date": due_date, "status": "draft"}
+
+def list_invoices(job_id=None):
+    conn = get_db()
+    if job_id:
+        rows = conn.execute("SELECT * FROM invoices WHERE job_id=? ORDER BY created_at DESC", (job_id,)).fetchall()
+    else:
+        rows = conn.execute("SELECT * FROM invoices ORDER BY created_at DESC").fetchall()
+    conn.close()
+    return {"invoices": rows_to_list(rows)}
+
+def get(invoice_id):
+    conn = get_db()
+    inv = row_to_dict(conn.execute("SELECT * FROM invoices WHERE id=?", (invoice_id,)).fetchone())
+    if not inv: conn.close(); return {"error": "invoice not found"}
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (inv["estimate_id"],)).fetchone())
+    if est:
+        inv["line_items"] = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (inv["estimate_id"],)).fetchall())
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (inv["job_id"],)).fetchone())
+    inv["job"] = job
+    conn.close()
+    return inv
+
+def mark_paid(invoice_id, amount_paid=None):
+    conn = get_db()
+    inv = row_to_dict(conn.execute("SELECT * FROM invoices WHERE id=?", (invoice_id,)).fetchone())
+    if not inv: conn.close(); return {"error": "invoice not found"}
+    paid = amount_paid if amount_paid else inv["amount_due"]
+    conn.execute("UPDATE invoices SET amount_paid=?, status='paid' WHERE id=?", (paid, invoice_id))
+    conn.commit(); conn.close()
+    return {"ok": True, "id": invoice_id, "amount_paid": paid, "status": "paid"}
+
+def sync_to_qbo(invoice_id):
+    """Sync invoice to QuickBooks Online."""
+    conn = get_db()
+    inv = row_to_dict(conn.execute("SELECT * FROM invoices WHERE id=?", (invoice_id,)).fetchone())
+    if not inv: conn.close(); return {"error": "invoice not found"}
+    items = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (inv["estimate_id"],)).fetchall())
+    conn.close()
+    try:
+        from lib.qbo import create_invoice as qbo_create
+        qbo_lines = [{"description": i["description"], "amount": i["total"], "qty": i["quantity"]} for i in items]
+        # Note: customer_id mapping needed — for now use placeholder
+        result = qbo_create("1", qbo_lines, inv["due_date"])
+        qbo_id = result.get("Invoice", {}).get("Id")
+        conn = get_db()
+        conn.execute("UPDATE invoices SET qbo_invoice_id=?, synced_to_qbo_at=? WHERE id=?", (qbo_id, now(), invoice_id))
+        conn.commit(); conn.close()
+        return {"ok": True, "qbo_invoice_id": qbo_id}
+    except Exception as e:
+        return {"error": str(e), "msg": "QBO sync failed — check token status"}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "list"
+    if cmd == "create": print(json.dumps(create(sys.argv[2], sys.argv[3] if len(sys.argv)>3 else None)))
+    elif cmd == "list": print(json.dumps(list_invoices(sys.argv[2] if len(sys.argv)>2 else None), indent=2))
+    elif cmd == "get": print(json.dumps(get(sys.argv[2]), indent=2))
+    elif cmd == "mark-paid": print(json.dumps(mark_paid(sys.argv[2], float(sys.argv[3]) if len(sys.argv)>3 else None)))
+    elif cmd == "sync-qbo": print(json.dumps(sync_to_qbo(sys.argv[2])))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/jobtread_sync.py b/scripts/jobtread_sync.py
new file mode 100644
index 0000000..8cce2d9
--- /dev/null
+++ b/scripts/jobtread_sync.py
@@ -0,0 +1,90 @@
+"""JobTread bidirectional sync tool."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+from lib.pave import pave_query, get_org_id
+
+def list_customers():
+    org_id = get_org_id()
+    result = pave_query({"organization": {"$": {"id": org_id},
+        "accounts": {"nodes": {"id": {}, "name": {}, "type": {}}}}})
+    accounts = result.get("organization", {}).get("accounts", {}).get("nodes", [])
+    return {"customers": [a for a in accounts if a.get("type") == "customer"]}
+
+def pull_jobs():
+    """Pull all jobs from JobTread and upsert to local DB."""
+    org_id = get_org_id()
+    result = pave_query({"organization": {"$": {"id": org_id},
+        "jobs": {"nodes": {"id": {}, "name": {}, "status": {}, "createdAt": {},
+                           "account": {"id": {}, "name": {}},
+                           "location": {"name": {}, "address": {}}}}}})
+    jobs = result.get("organization", {}).get("jobs", {}).get("nodes", [])
+    conn = get_db()
+    synced = 0
+    for j in jobs:
+        jt_id = j.get("id")
+        existing = conn.execute("SELECT id FROM jobs WHERE jobtread_job_id=?", (jt_id,)).fetchone()
+        if existing:
+            conn.execute("UPDATE jobs SET status=?, updated_at=?, synced_to_jobtread_at=? WHERE jobtread_job_id=?",
+                (j.get("status","active"), now(), now(), jt_id))
+        else:
+            conn.execute("""INSERT INTO jobs (id, client_name, address, description, jobtread_job_id, jobtread_customer_id, status, created_at, updated_at, synced_to_jobtread_at)
+                VALUES (?,?,?,?,?,?,?,?,?,?)""",
+                (new_id(), j.get("account",{}).get("name",""), j.get("location",{}).get("address",""),
+                 j.get("name",""), jt_id, j.get("account",{}).get("id",""),
+                 j.get("status","active"), j.get("createdAt", now()), now(), now()))
+        synced += 1
+    conn.commit(); conn.close()
+    _log_sync(None, None, "job", "pull", "success")
+    return {"ok": True, "synced": synced, "total_jobs": len(jobs)}
+
+def push_estimate(estimate_id):
+    """Push local estimate to JobTread. NOTE: Pave create mutations need verification."""
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (est["job_id"],)).fetchone())
+    if not job or not job.get("jobtread_job_id"):
+        conn.close(); return {"error": "job not linked to JobTread"}
+    items = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (estimate_id,)).fetchall())
+    conn.close()
+    # Log as pending — actual Pave create mutation TBD (needs Pave Explorer verification)
+    _log_sync(estimate_id, job["jobtread_job_id"], "estimate", "push", "pending",
+              "Pave create document mutation needs verification")
+    return {"ok": True, "status": "pending", "msg": "Estimate queued for JobTread sync. Pave create mutation needs verification via Pave Explorer."}
+
+def push_invoice(invoice_id):
+    """Push local invoice to JobTread."""
+    conn = get_db()
+    inv = row_to_dict(conn.execute("SELECT * FROM invoices WHERE id=?", (invoice_id,)).fetchone())
+    if not inv: conn.close(); return {"error": "invoice not found"}
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (inv["job_id"],)).fetchone())
+    conn.close()
+    if not job or not job.get("jobtread_job_id"):
+        return {"error": "job not linked to JobTread"}
+    _log_sync(invoice_id, job["jobtread_job_id"], "invoice", "push", "pending",
+              "Pave create invoice mutation needs verification")
+    return {"ok": True, "status": "pending", "msg": "Invoice queued for JobTread sync."}
+
+def get_sync_status(job_id):
+    conn = get_db()
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (job_id,)).fetchone())
+    logs = rows_to_list(conn.execute("SELECT * FROM jobtread_sync_log WHERE local_id=? ORDER BY synced_at DESC LIMIT 10", (job_id,)).fetchall())
+    conn.close()
+    return {"job": job, "sync_log": logs}
+
+def _log_sync(local_id, jt_id, entity_type, action, status, error=None):
+    conn = get_db()
+    conn.execute("INSERT INTO jobtread_sync_log (id, local_id, jobtread_id, entity_type, action, status, error_message, synced_at) VALUES (?,?,?,?,?,?,?,?)",
+        (new_id(), local_id, jt_id, entity_type, action, status, error, now()))
+    conn.commit(); conn.close()
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "list-customers"
+    if cmd == "list-customers": print(json.dumps(list_customers(), indent=2))
+    elif cmd == "pull-jobs": print(json.dumps(pull_jobs(), indent=2))
+    elif cmd == "push-estimate": print(json.dumps(push_estimate(sys.argv[2])))
+    elif cmd == "push-invoice": print(json.dumps(push_invoice(sys.argv[2])))
+    elif cmd == "sync-status": print(json.dumps(get_sync_status(sys.argv[2]), indent=2))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/material_list.py b/scripts/material_list.py
new file mode 100644
index 0000000..d2f57e9
--- /dev/null
+++ b/scripts/material_list.py
@@ -0,0 +1,35 @@
+"""Material list generator from estimates."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, row_to_dict, rows_to_list
+
+def generate(estimate_id):
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    items = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=? AND category='materials'", (estimate_id,)).fetchall())
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (est["job_id"],)).fetchone())
+    conn.close()
+
+    materials = []
+    total = 0
+    for item in items:
+        materials.append({
+            "item": item["description"],
+            "quantity": item["quantity"],
+            "unit": item["unit"],
+            "est_cost": round(item["total"], 2)
+        })
+        total += item["total"]
+
+    return {
+        "job": job.get("description", "") if job else "",
+        "materials": materials,
+        "total_materials_cost": round(total, 2),
+        "items_count": len(materials)
+    }
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    if len(sys.argv) < 2: print(json.dumps({"error": "usage: material_list.py <estimate_id>"}))
+    else: print(json.dumps(generate(sys.argv[1]), indent=2))
diff --git a/scripts/pricing_catalog.py b/scripts/pricing_catalog.py
new file mode 100644
index 0000000..178988f
--- /dev/null
+++ b/scripts/pricing_catalog.py
@@ -0,0 +1,71 @@
+"""Pricing catalog CRUD operations."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+
+def list_catalogs():
+    conn = get_db()
+    rows = conn.execute("SELECT * FROM pricing_catalogs ORDER BY created_at DESC").fetchall()
+    conn.close()
+    result = rows_to_list(rows)
+    for r in result:
+        r["labor_rates"] = json.loads(r["labor_rates"]) if r["labor_rates"] else {}
+        r["material_markups"] = json.loads(r["material_markups"]) if r["material_markups"] else {}
+    return {"catalogs": result}
+
+def create_catalog(name, labor_rates, material_markups=None):
+    cid = new_id()
+    conn = get_db()
+    conn.execute("INSERT INTO pricing_catalogs (id, name, labor_rates, material_markups, created_at) VALUES (?,?,?,?,?)",
+        (cid, name, json.dumps(labor_rates), json.dumps(material_markups or {}), now()))
+    conn.commit(); conn.close()
+    return {"id": cid, "name": name}
+
+def get_catalog(catalog_id):
+    conn = get_db()
+    row = conn.execute("SELECT * FROM pricing_catalogs WHERE id=?", (catalog_id,)).fetchone()
+    conn.close()
+    if not row: return {"error": "not found"}
+    r = row_to_dict(row)
+    r["labor_rates"] = json.loads(r["labor_rates"]) if r["labor_rates"] else {}
+    r["material_markups"] = json.loads(r["material_markups"]) if r["material_markups"] else {}
+    return r
+
+def update_catalog(catalog_id, name=None, labor_rates=None, material_markups=None):
+    conn = get_db()
+    if name: conn.execute("UPDATE pricing_catalogs SET name=? WHERE id=?", (name, catalog_id))
+    if labor_rates: conn.execute("UPDATE pricing_catalogs SET labor_rates=? WHERE id=?", (json.dumps(labor_rates), catalog_id))
+    if material_markups: conn.execute("UPDATE pricing_catalogs SET material_markups=? WHERE id=?", (json.dumps(material_markups), catalog_id))
+    conn.commit(); conn.close()
+    return {"ok": True, "id": catalog_id}
+
+def delete_catalog(catalog_id):
+    conn = get_db()
+    conn.execute("DELETE FROM pricing_catalogs WHERE id=?", (catalog_id,))
+    conn.commit(); conn.close()
+    return {"ok": True, "deleted": catalog_id}
+
+def seed_default():
+    """Seed default Kellow Construction pricing catalog."""
+    existing = list_catalogs()
+    if existing["catalogs"]: return {"ok": True, "msg": "already seeded"}
+    return create_catalog("Kellow Default", {
+        "general_labor": 55, "framing": 85, "electrical": 95,
+        "plumbing": 90, "painting": 65, "drywall": 75,
+        "tile": 80, "flooring": 70, "roofing": 85,
+        "concrete": 75, "demolition": 60, "finish_carpentry": 90,
+        "project_management": 95, "design": 110
+    }, {"lumber": 1.15, "drywall": 1.20, "tile": 1.25,
+        "plumbing_fixtures": 1.20, "electrical_fixtures": 1.15,
+        "paint": 1.10, "hardware": 1.15, "appliances": 1.10})
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "list"
+    if cmd == "list": print(json.dumps(list_catalogs(), indent=2))
+    elif cmd == "create": print(json.dumps(create_catalog(sys.argv[2], json.loads(sys.argv[3]), json.loads(sys.argv[4]) if len(sys.argv) > 4 else None)))
+    elif cmd == "get": print(json.dumps(get_catalog(sys.argv[2]), indent=2))
+    elif cmd == "update": print(json.dumps(update_catalog(sys.argv[2], labor_rates=json.loads(sys.argv[3]) if len(sys.argv) > 3 else None)))
+    elif cmd == "delete": print(json.dumps(delete_catalog(sys.argv[2])))
+    elif cmd == "seed": print(json.dumps(seed_default()))
+    else: print(json.dumps({"error": f"unknown command: {cmd}"}))
diff --git a/scripts/project_mgmt.py b/scripts/project_mgmt.py
new file mode 100644
index 0000000..f3f5cc4
--- /dev/null
+++ b/scripts/project_mgmt.py
@@ -0,0 +1,102 @@
+"""Project management: jobs and estimates CRUD."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+
+def create_job(client_name, address=None, description=None, jobtread_job_id=None, jobtread_customer_id=None):
+    jid = new_id()
+    conn = get_db()
+    conn.execute("""INSERT INTO jobs (id, client_name, address, description, jobtread_job_id, jobtread_customer_id, status, created_at, updated_at)
+        VALUES (?,?,?,?,?,?,?,?,?)""",
+        (jid, client_name, address, description, jobtread_job_id, jobtread_customer_id, "proposal", now(), now()))
+    conn.commit(); conn.close()
+    return {"id": jid, "client_name": client_name, "status": "proposal"}
+
+def list_jobs(status=None):
+    conn = get_db()
+    if status:
+        rows = conn.execute("SELECT * FROM jobs WHERE status=? ORDER BY created_at DESC", (status,)).fetchall()
+    else:
+        rows = conn.execute("SELECT * FROM jobs ORDER BY created_at DESC").fetchall()
+    conn.close()
+    return {"jobs": rows_to_list(rows)}
+
+def get_job(job_id):
+    conn = get_db()
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (job_id,)).fetchone())
+    if not job: conn.close(); return {"error": "job not found"}
+    job["estimates"] = rows_to_list(conn.execute("SELECT * FROM estimates WHERE job_id=?", (job_id,)).fetchall())
+    job["invoices"] = rows_to_list(conn.execute("SELECT * FROM invoices WHERE job_id=?", (job_id,)).fetchall())
+    conn.close()
+    return job
+
+def update_job(job_id, status=None, client_name=None, address=None, description=None):
+    conn = get_db()
+    updates = []
+    params = []
+    for field, val in [("status", status), ("client_name", client_name), ("address", address), ("description", description)]:
+        if val is not None:
+            updates.append(f"{field}=?"); params.append(val)
+    if updates:
+        updates.append("updated_at=?"); params.append(now())
+        params.append(job_id)
+        conn.execute(f"UPDATE jobs SET {','.join(updates)} WHERE id=?", params)
+        conn.commit()
+    conn.close()
+    return {"ok": True, "id": job_id}
+
+def create_estimate(job_id, line_items, labor_hours=None, labor_rate=None, markup_percent=20):
+    eid = new_id()
+    conn = get_db()
+    materials_cost = 0
+    total_labor_hours = labor_hours or 0
+    # Calculate totals first
+    item_rows = []
+    for item in line_items:
+        lid = new_id()
+        total = item.get("quantity", 1) * item.get("unit_cost", 0)
+        item_rows.append((lid, eid, item.get("description",""), item.get("category",""), item.get("quantity",1), item.get("unit","ea"), item.get("unit_cost",0), total))
+        if item.get("category") == "labor":
+            total_labor_hours += item.get("quantity", 0)
+        else:
+            materials_cost += total
+    effective_rate = labor_rate or 85
+    labor_cost = total_labor_hours * effective_rate
+    subtotal = labor_cost + materials_cost
+    markup = subtotal * (markup_percent / 100)
+    total_cost = subtotal + markup
+    # Insert estimate first, then line items
+    conn.execute("""INSERT INTO estimates (id, job_id, labor_hours, labor_rate, materials_cost, markup_percent, total_cost, status, created_at)
+        VALUES (?,?,?,?,?,?,?,?,?)""",
+        (eid, job_id, total_labor_hours, effective_rate, materials_cost, markup_percent, total_cost, "draft", now()))
+    for row in item_rows:
+        conn.execute("INSERT INTO line_items (id, estimate_id, description, category, quantity, unit, unit_cost, total) VALUES (?,?,?,?,?,?,?,?)", row)
+    conn.commit(); conn.close()
+    return {"id": eid, "job_id": job_id, "labor_hours": total_labor_hours, "labor_rate": effective_rate,
+            "materials_cost": materials_cost, "markup_percent": markup_percent, "total_cost": total_cost}
+
+def get_estimate(estimate_id):
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    est["line_items"] = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (estimate_id,)).fetchall())
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (est["job_id"],)).fetchone())
+    est["job"] = job
+    conn.close()
+    return est
+
+def update_estimate_status(estimate_id, status):
+    conn = get_db()
+    conn.execute("UPDATE estimates SET status=? WHERE id=?", (status, estimate_id))
+    conn.commit(); conn.close()
+    return {"ok": True, "id": estimate_id, "status": status}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "list-jobs"
+    if cmd == "create-job": print(json.dumps(create_job(sys.argv[2], sys.argv[3] if len(sys.argv)>3 else None, sys.argv[4] if len(sys.argv)>4 else None)))
+    elif cmd == "list-jobs": print(json.dumps(list_jobs(sys.argv[2] if len(sys.argv)>2 else None), indent=2))
+    elif cmd == "get-job": print(json.dumps(get_job(sys.argv[2]), indent=2))
+    elif cmd == "get-estimate": print(json.dumps(get_estimate(sys.argv[2]), indent=2))
+    elif cmd == "create-estimate": print(json.dumps(create_estimate(sys.argv[2], json.loads(sys.argv[3]), markup_percent=float(sys.argv[4]) if len(sys.argv)>4 else 20)))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/project_scheduler.py b/scripts/project_scheduler.py
new file mode 100644
index 0000000..7b625a9
--- /dev/null
+++ b/scripts/project_scheduler.py
@@ -0,0 +1,35 @@
+"""Project scheduler: milestones and timeline management."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, row_to_dict, rows_to_list
+
+def set_milestone(job_id, phase_name, start_date=None, end_date=None, status="pending"):
+    mid = new_id()
+    conn = get_db()
+    conn.execute("INSERT INTO project_milestones (id, job_id, phase_name, start_date, end_date, status) VALUES (?,?,?,?,?,?)",
+        (mid, job_id, phase_name, start_date, end_date, status))
+    conn.commit(); conn.close()
+    return {"id": mid, "job_id": job_id, "phase": phase_name}
+
+def update_milestone(milestone_id, status=None, start_date=None, end_date=None):
+    conn = get_db()
+    if status: conn.execute("UPDATE project_milestones SET status=? WHERE id=?", (status, milestone_id))
+    if start_date: conn.execute("UPDATE project_milestones SET start_date=? WHERE id=?", (start_date, milestone_id))
+    if end_date: conn.execute("UPDATE project_milestones SET end_date=? WHERE id=?", (end_date, milestone_id))
+    conn.commit(); conn.close()
+    return {"ok": True, "id": milestone_id}
+
+def get_schedule(job_id):
+    conn = get_db()
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (job_id,)).fetchone())
+    milestones = rows_to_list(conn.execute("SELECT * FROM project_milestones WHERE job_id=? ORDER BY start_date", (job_id,)).fetchall())
+    conn.close()
+    if not job: return {"error": "job not found"}
+    return {"job_id": job_id, "client": job.get("client_name"), "milestones": milestones}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "get"
+    if cmd == "set": print(json.dumps(set_milestone(sys.argv[2], sys.argv[3], sys.argv[4] if len(sys.argv)>4 else None, sys.argv[5] if len(sys.argv)>5 else None)))
+    elif cmd == "get": print(json.dumps(get_schedule(sys.argv[2]), indent=2))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/proposal_drafter.py b/scripts/proposal_drafter.py
new file mode 100644
index 0000000..18c6830
--- /dev/null
+++ b/scripts/proposal_drafter.py
@@ -0,0 +1,104 @@
+"""Proposal drafter: generate professional proposals from estimates."""
+import json, sys, os
+from datetime import datetime, timedelta
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, row_to_dict, rows_to_list
+
+TEMPLATE = """# Proposal — {client_name}
+## {job_description}
+
+**Prepared by:** Kellow Construction
+**Date:** {date}
+**Valid for:** 30 days
+
+---
+
+## Scope of Work
+
+{scope_items}
+
+## Investment Summary
+
+| Category | Amount |
+|----------|--------|
+| Labor ({labor_hours} hrs @ ${labor_rate}/hr) | ${labor_cost:,.2f} |
+| Materials | ${materials_cost:,.2f} |
+| Subtotal | ${subtotal:,.2f} |
+| Overhead & Profit ({markup_percent}%) | ${markup:,.2f} |
+| **Total** | **${total_cost:,.2f}** |
+
+## Timeline
+
+Estimated duration: {timeline}
+Projected start: Upon approval and permitting
+
+## Payment Terms
+
+- 30% deposit upon contract signing
+- Progress payments at milestones
+- Final 10% upon completion and walkthrough
+
+## Exclusions
+
+- Permits and fees (handled separately)
+- Unforeseen structural issues
+- Owner-supplied materials
+- Landscaping and exterior work (unless specified)
+
+## Warranty
+
+Kellow Construction provides a 1-year workmanship warranty on all labor performed.
+
+---
+
+*This proposal is valid for 30 days from the date above. Acceptance constitutes agreement to the terms outlined herein.*
+"""
+
+def draft(estimate_id):
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    items = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (estimate_id,)).fetchall())
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (est["job_id"],)).fetchone())
+    conn.close()
+
+    labor_cost = est["labor_hours"] * est["labor_rate"]
+    subtotal = labor_cost + est["materials_cost"]
+    markup = subtotal * (est["markup_percent"] / 100)
+
+    # Build scope items
+    scope_lines = []
+    for item in items:
+        scope_lines.append(f"- {item['description']}: {item['quantity']} {item['unit']} @ ${item['unit_cost']:.2f}/{item['unit']}")
+
+    # Estimate timeline from labor hours
+    weeks = max(1, int(est["labor_hours"] / 40))
+    timeline = f"{weeks} week{'s' if weeks > 1 else ''}"
+
+    proposal = TEMPLATE.format(
+        client_name=job.get("client_name", "Client") if job else "Client",
+        job_description=job.get("description", "Project") if job else "Project",
+        date=datetime.now().strftime("%B %d, %Y"),
+        scope_items="\n".join(scope_lines) if scope_lines else "- As discussed",
+        labor_hours=round(est["labor_hours"], 1),
+        labor_rate=est["labor_rate"],
+        labor_cost=labor_cost,
+        materials_cost=est["materials_cost"],
+        subtotal=subtotal,
+        markup_percent=est["markup_percent"],
+        markup=markup,
+        total_cost=est["total_cost"],
+        timeline=timeline,
+    )
+    return {"proposal": proposal, "estimate_id": estimate_id, "total": est["total_cost"]}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    if len(sys.argv) < 2:
+        print(json.dumps({"error": "usage: proposal_drafter.py <estimate_id>"}))
+    else:
+        result = draft(sys.argv[1])
+        if "proposal" in result:
+            print(result["proposal"])
+        else:
+            print(json.dumps(result))
diff --git a/scripts/scope_generator.py b/scripts/scope_generator.py
new file mode 100644
index 0000000..b4d4e93
--- /dev/null
+++ b/scripts/scope_generator.py
@@ -0,0 +1,50 @@
+"""Scope of work generator from estimates."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, row_to_dict, rows_to_list
+
+STANDARD_CLAUSES = """
+## General Conditions
+
+- All work performed in accordance with local building codes and regulations
+- Contractor to obtain all necessary permits unless otherwise noted
+- Work area to be kept clean and debris removed daily
+- Final cleanup included upon project completion
+
+## Warranty
+
+- 1-year workmanship warranty on all labor
+- Manufacturer warranties apply to all materials and fixtures
+
+## Exclusions
+
+- Unforeseen conditions (mold, asbestos, structural damage) — addressed via change order
+- Owner-supplied materials — contractor not responsible for defects
+- Landscaping, exterior painting, or work not explicitly listed above
+- Furniture moving or storage
+"""
+
+def generate(estimate_id):
+    conn = get_db()
+    est = row_to_dict(conn.execute("SELECT * FROM estimates WHERE id=?", (estimate_id,)).fetchone())
+    if not est: conn.close(); return {"error": "estimate not found"}
+    items = rows_to_list(conn.execute("SELECT * FROM line_items WHERE estimate_id=?", (estimate_id,)).fetchall())
+    job = row_to_dict(conn.execute("SELECT * FROM jobs WHERE id=?", (est["job_id"],)).fetchone())
+    conn.close()
+
+    lines = [f"# Scope of Work — {job.get('client_name','Client') if job else 'Client'}",
+             f"**Project:** {job.get('description','') if job else ''}",
+             f"**Address:** {job.get('address','') if job else ''}", "", "## Work Items", ""]
+    for i, item in enumerate(items, 1):
+        if item["category"] == "labor":
+            lines.append(f"{i}. **{item['description']}** — {item['quantity']} {item['unit']}")
+    lines.append(STANDARD_CLAUSES)
+    scope = "\n".join(lines)
+    return {"scope": scope, "estimate_id": estimate_id, "items_count": len(items)}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    if len(sys.argv) < 2: print(json.dumps({"error": "usage: scope_generator.py <estimate_id>"}))
+    else:
+        r = generate(sys.argv[1])
+        print(r.get("scope", json.dumps(r)))
diff --git a/scripts/signature_mgmt.py b/scripts/signature_mgmt.py
new file mode 100644
index 0000000..4624ee8
--- /dev/null
+++ b/scripts/signature_mgmt.py
@@ -0,0 +1,35 @@
+"""Signature management: request and record client signatures."""
+import json, sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import get_db, new_id, now, row_to_dict, rows_to_list
+
+def request_signature(estimate_id, client_name):
+    sid = new_id()
+    conn = get_db()
+    conn.execute("INSERT INTO signatures (id, estimate_id, client_name) VALUES (?,?,?)",
+        (sid, estimate_id, client_name))
+    conn.commit(); conn.close()
+    return {"signature_id": sid, "estimate_id": estimate_id, "client_name": client_name, "status": "pending"}
+
+def record_signature(signature_id, signature_data="approved_via_telegram"):
+    conn = get_db()
+    conn.execute("UPDATE signatures SET signature_data=?, signed_at=? WHERE id=?",
+        (signature_data, now(), signature_id))
+    conn.commit(); conn.close()
+    # Also log as activity
+    sig = row_to_dict(conn.execute("SELECT * FROM signatures WHERE id=?", (signature_id,)).fetchone()) if False else None
+    return {"ok": True, "signature_id": signature_id, "signed_at": now()}
+
+def get_signature(estimate_id):
+    conn = get_db()
+    row = row_to_dict(conn.execute("SELECT * FROM signatures WHERE estimate_id=? ORDER BY signed_at DESC LIMIT 1", (estimate_id,)).fetchone())
+    conn.close()
+    return row if row else {"status": "no signature found"}
+
+if __name__ == "__main__":
+    from lib.db import init_db; init_db()
+    cmd = sys.argv[1] if len(sys.argv) > 1 else "get"
+    if cmd == "request": print(json.dumps(request_signature(sys.argv[2], sys.argv[3])))
+    elif cmd == "record": print(json.dumps(record_signature(sys.argv[2], sys.argv[3] if len(sys.argv)>3 else "approved")))
+    elif cmd == "get": print(json.dumps(get_signature(sys.argv[2]), indent=2))
+    else: print(json.dumps({"error": f"unknown: {cmd}"}))
diff --git a/scripts/site_analyzer.py b/scripts/site_analyzer.py
new file mode 100644
index 0000000..c7503ed
--- /dev/null
+++ b/scripts/site_analyzer.py
@@ -0,0 +1,44 @@
+"""Site walkthrough analyzer: structured report from observations."""
+import json, sys, os, re
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+from lib.db import now
+
+# Issue patterns to flag
+ISSUE_PATTERNS = {
+    r"water\s*damage|moisture|leak|mold": {"severity": "high", "category": "water_damage", "cost_impact": 1200},
+    r"crack|structural|foundation\s*issue|settling": {"severity": "high", "category": "structural", "cost_impact": 3000},
+    r"rot|decay|termite|pest": {"severity": "high", "category": "pest_damage", "cost_impact": 2000},
+    r"asbestos|lead\s*paint|hazmat": {"severity": "critical", "category": "hazmat", "cost_impact": 5000},
+    r"outdated\s*wiring|knob.and.tube|electrical\s*issue": {"severity": "medium", "category": "electrical", "cost_impact": 2500},
+    r"rust|corrosion|pipe": {"severity": "medium", "category": "plumbing", "cost_impact": 1500},
+}
+
+def analyze(observations, job_id=None, photos_description=None):
+    """Analyze site observations and flag issues."""
+    combined = f"{observations} {photos_description or ''}"
+    issues = []
+    total_impact = 0
+    for pattern, info in ISSUE_PATTERNS.items():
+        if re.search(pattern, combined, re.I):
+            issues.append({
+                "category": info["category"],
+                "severity": info["severity"],
+                "estimated_cost_impact": info["cost_impact"],
+                "matched": re.search(pattern, combined, re.I).group(0)
+            })
+            total_impact += info["cost_impact"]
+
+    return {
+        "job_id": job_id,
+        "analyzed_at": now(),
+        "observations": observations,
+        "photos_description": photos_description,
+        "issues_found": len(issues),
+        "issues": issues,
+        "total_cost_impact": total_impact,
+        "recommendation": f"Add ${total_impact:,.0f} to estimate for remediation" if issues else "No significant issues detected"
+    }
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2: print(json.dumps({"error": "usage: site_analyzer.py <observations> [photos_desc]"}))
+    else: print(json.dumps(analyze(sys.argv[1], photos_description=sys.argv[2] if len(sys.argv)>2 else None), indent=2))
diff --git a/server.py b/server.py
new file mode 100644
index 0000000..7b359a1
--- /dev/null
+++ b/server.py
@@ -0,0 +1,343 @@
+"""Handoff Pro MCP Server — SSE transport for mcporter compatibility."""
+import json
+import os
+import sys
+import queue
+import threading
+import uuid
+from http.server import HTTPServer, BaseHTTPRequestHandler
+from urllib.parse import urlparse, parse_qs
+
+sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
+from lib import db
+
+# Import tool handlers
+from scripts import pricing_catalog, project_mgmt, jobtread_sync
+from scripts import estimate_generator, proposal_drafter, invoice_mgmt
+from scripts import scope_generator, material_list, daily_log
+from scripts import site_analyzer, budget_tracker, project_scheduler
+from scripts import signature_mgmt, activity_logger
+
+PORT = int(os.environ.get("HANDOFF_PORT", 3101))
+
+# Tool registry: name -> {description, input_schema, handler}
+TOOLS = {}
+
+def tool(name, description, schema):
+    """Decorator to register an MCP tool."""
+    def decorator(fn):
+        TOOLS[name] = {"description": description, "inputSchema": schema, "handler": fn}
+        return fn
+    return decorator
+
+# --- Pricing Catalog Tools ---
+@tool("list_catalogs", "List all pricing catalogs", {"type": "object", "properties": {}})
+def _list_catalogs(args): return pricing_catalog.list_catalogs()
+
+@tool("create_catalog", "Create a pricing catalog", {
+    "type": "object", "properties": {
+        "name": {"type": "string"}, "labor_rates": {"type": "object"},
+        "material_markups": {"type": "object"}
+    }, "required": ["name", "labor_rates"]})
+def _create_catalog(args): return pricing_catalog.create_catalog(**args)
+
+@tool("get_catalog", "Get a pricing catalog by ID", {
+    "type": "object", "properties": {"catalog_id": {"type": "string"}}, "required": ["catalog_id"]})
+def _get_catalog(args): return pricing_catalog.get_catalog(args["catalog_id"])
+
+@tool("update_catalog", "Update a pricing catalog", {
+    "type": "object", "properties": {
+        "catalog_id": {"type": "string"}, "name": {"type": "string"},
+        "labor_rates": {"type": "object"}, "material_markups": {"type": "object"}
+    }, "required": ["catalog_id"]})
+def _update_catalog(args): return pricing_catalog.update_catalog(**args)
+
+@tool("delete_catalog", "Delete a pricing catalog", {
+    "type": "object", "properties": {"catalog_id": {"type": "string"}}, "required": ["catalog_id"]})
+def _delete_catalog(args): return pricing_catalog.delete_catalog(args["catalog_id"])
+
+# --- Project Management Tools ---
+@tool("create_job", "Create a new job", {
+    "type": "object", "properties": {
+        "client_name": {"type": "string"}, "address": {"type": "string"},
+        "description": {"type": "string"}
+    }, "required": ["client_name"]})
+def _create_job(args): return project_mgmt.create_job(**args)
+
+@tool("list_jobs", "List all jobs", {
+    "type": "object", "properties": {"status": {"type": "string"}}})
+def _list_jobs(args): return project_mgmt.list_jobs(args.get("status"))
+
+@tool("get_job", "Get job details", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}, "required": ["job_id"]})
+def _get_job(args): return project_mgmt.get_job(args["job_id"])
+
+@tool("update_job", "Update a job", {
+    "type": "object", "properties": {
+        "job_id": {"type": "string"}, "status": {"type": "string"},
+        "client_name": {"type": "string"}, "address": {"type": "string"},
+        "description": {"type": "string"}
+    }, "required": ["job_id"]})
+def _update_job(args): return project_mgmt.update_job(**args)
+
+@tool("create_estimate", "Create an estimate for a job", {
+    "type": "object", "properties": {
+        "job_id": {"type": "string"},
+        "line_items": {"type": "array", "items": {"type": "object"}},
+        "labor_hours": {"type": "number"}, "labor_rate": {"type": "number"},
+        "markup_percent": {"type": "number"}
+    }, "required": ["job_id", "line_items"]})
+def _create_estimate(args): return project_mgmt.create_estimate(**args)
+
+@tool("get_estimate", "Get estimate details with line items", {
+    "type": "object", "properties": {"estimate_id": {"type": "string"}}, "required": ["estimate_id"]})
+def _get_estimate(args): return project_mgmt.get_estimate(args["estimate_id"])
+
+# --- JobTread Sync Tools ---
+@tool("jt_list_customers", "List customers from JobTread", {"type": "object", "properties": {}})
+def _jt_customers(args): return jobtread_sync.list_customers()
+
+@tool("jt_pull_jobs", "Pull all jobs from JobTread into local DB", {"type": "object", "properties": {}})
+def _jt_pull(args): return jobtread_sync.pull_jobs()
+
+@tool("jt_push_estimate", "Push estimate to JobTread", {
+    "type": "object", "properties": {"estimate_id": {"type": "string"}}, "required": ["estimate_id"]})
+def _jt_push_est(args): return jobtread_sync.push_estimate(args["estimate_id"])
+
+@tool("jt_push_invoice", "Push invoice to JobTread", {
+    "type": "object", "properties": {"invoice_id": {"type": "string"}}, "required": ["invoice_id"]})
+def _jt_push_inv(args): return jobtread_sync.push_invoice(args["invoice_id"])
+
+@tool("jt_sync_status", "Get sync status for a job", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}, "required": ["job_id"]})
+def _jt_status(args): return jobtread_sync.get_sync_status(args["job_id"])
+
+# --- Estimate Generator ---
+@tool("generate_estimate", "Generate estimate from job description", {
+    "type": "object", "properties": {
+        "description": {"type": "string"}, "job_id": {"type": "string"},
+        "catalog_id": {"type": "string"}, "markup_percent": {"type": "number"}
+    }, "required": ["description"]})
+def _gen_estimate(args): return estimate_generator.generate(**args)
+
+# --- Proposal Drafter ---
+@tool("draft_proposal", "Generate a professional proposal from an estimate", {
+    "type": "object", "properties": {"estimate_id": {"type": "string"}}, "required": ["estimate_id"]})
+def _draft_proposal(args): return proposal_drafter.draft(args["estimate_id"])
+
+# --- Invoice Management ---
+@tool("create_invoice", "Create invoice from estimate", {
+    "type": "object", "properties": {
+        "estimate_id": {"type": "string"}, "due_date": {"type": "string"}
+    }, "required": ["estimate_id"]})
+def _create_invoice(args): return invoice_mgmt.create(**args)
+
+@tool("list_invoices", "List invoices", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}})
+def _list_invoices(args): return invoice_mgmt.list_invoices(args.get("job_id"))
+
+@tool("mark_invoice_paid", "Mark invoice as paid", {
+    "type": "object", "properties": {
+        "invoice_id": {"type": "string"}, "amount_paid": {"type": "number"}
+    }, "required": ["invoice_id"]})
+def _mark_paid(args): return invoice_mgmt.mark_paid(**args)
+
+@tool("sync_invoice_qbo", "Sync invoice to QuickBooks Online", {
+    "type": "object", "properties": {"invoice_id": {"type": "string"}}, "required": ["invoice_id"]})
+def _sync_qbo(args): return invoice_mgmt.sync_to_qbo(args["invoice_id"])
+
+# --- Scope & Materials ---
+@tool("generate_scope", "Generate scope of work from estimate", {
+    "type": "object", "properties": {"estimate_id": {"type": "string"}}, "required": ["estimate_id"]})
+def _gen_scope(args): return scope_generator.generate(args["estimate_id"])
+
+@tool("generate_material_list", "Generate material shopping list from estimate", {
+    "type": "object", "properties": {"estimate_id": {"type": "string"}}, "required": ["estimate_id"]})
+def _gen_materials(args): return material_list.generate(args["estimate_id"])
+
+# --- Daily Logs ---
+@tool("create_daily_log", "Create a daily job log entry", {
+    "type": "object", "properties": {
+        "job_id": {"type": "string"}, "crew_names": {"type": "string"},
+        "hours_worked": {"type": "number"}, "notes": {"type": "string"},
+        "photos": {"type": "array", "items": {"type": "string"}}
+    }, "required": ["job_id", "notes"]})
+def _create_log(args): return daily_log.create(**args)
+
+@tool("list_daily_logs", "List daily logs for a job", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}, "required": ["job_id"]})
+def _list_logs(args): return daily_log.list_logs(args["job_id"])
+
+# --- Site Analyzer ---
+@tool("analyze_site", "Analyze site conditions from observations", {
+    "type": "object", "properties": {
+        "job_id": {"type": "string"}, "observations": {"type": "string"},
+        "photos_description": {"type": "string"}
+    }, "required": ["observations"]})
+def _analyze_site(args): return site_analyzer.analyze(**args)
+
+# --- Budget Tracker ---
+@tool("get_budget_report", "Get budget vs actual report for a job", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}, "required": ["job_id"]})
+def _budget_report(args): return budget_tracker.report(args["job_id"])
+
+# --- Project Scheduler ---
+@tool("set_milestone", "Set a project milestone", {
+    "type": "object", "properties": {
+        "job_id": {"type": "string"}, "phase_name": {"type": "string"},
+        "start_date": {"type": "string"}, "end_date": {"type": "string"}
+    }, "required": ["job_id", "phase_name"]})
+def _set_milestone(args): return project_scheduler.set_milestone(**args)
+
+@tool("get_schedule", "Get project schedule/timeline", {
+    "type": "object", "properties": {"job_id": {"type": "string"}}, "required": ["job_id"]})
+def _get_schedule(args): return project_scheduler.get_schedule(args["job_id"])
+
+# --- Signatures & Activity ---
+@tool("request_signature", "Request client signature on estimate", {
+    "type": "object", "properties": {
+        "estimate_id": {"type": "string"}, "client_name": {"type": "string"}
+    }, "required": ["estimate_id", "client_name"]})
+def _req_sig(args): return signature_mgmt.request_signature(**args)
+
+@tool("record_signature", "Record a client signature", {
+    "type": "object", "properties": {
+        "signature_id": {"type": "string"}, "signature_data": {"type": "string"}
+    }, "required": ["signature_id"]})
+def _rec_sig(args): return signature_mgmt.record_signature(**args)
+
+@tool("log_activity", "Log client activity (viewed, approved, signed, paid)", {
+    "type": "object", "properties": {
+        "estimate_id": {"type": "string"}, "job_id": {"type": "string"},
+        "action": {"type": "string"}
+    }, "required": ["action"]})
+def _log_activity(args): return activity_logger.log(**args)
+
+@tool("get_activity", "Get activity log for an estimate or job", {
+    "type": "object", "properties": {
+        "estimate_id": {"type": "string"}, "job_id": {"type": "string"}
+    }})
+def _get_activity(args): return activity_logger.get(**args)
+
+# --- MCP SSE Server Implementation ---
+class SSEClient:
+    def __init__(self):
+        self.queue = queue.Queue()
+        self.id = str(uuid.uuid4())
+
+clients = {}
+
+def send_sse(client_id, event, data):
+    if client_id in clients:
+        clients[client_id].queue.put(f"event: {event}\ndata: {json.dumps(data)}\n\n")
+
+class MCPHandler(BaseHTTPRequestHandler):
+    def log_message(self, format, *args): pass
+
+    def do_GET(self):
+        path = urlparse(self.path).path
+        if path == "/sse":
+            self._handle_sse()
+        elif path == "/health":
+            self._json_response({"ok": True, "tools": len(TOOLS)})
+        else:
+            self.send_error(404)
+
+    def do_POST(self):
+        path = urlparse(self.path).path
+        if path == "/messages":
+            self._handle_message()
+        else:
+            self.send_error(404)
+
+    def _handle_sse(self):
+        client = SSEClient()
+        clients[client.id] = client
+        self.send_response(200)
+        self.send_header("Content-Type", "text/event-stream")
+        self.send_header("Cache-Control", "no-cache")
+        self.send_header("Connection", "keep-alive")
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.end_headers()
+        # Send endpoint info
+        endpoint_msg = {"jsonrpc": "2.0", "method": "endpoint",
+                        "params": {"uri": f"http://0.0.0.0:{PORT}/messages?client_id={client.id}"}}
+        self.wfile.write(f"event: endpoint\ndata: {json.dumps(endpoint_msg)}\n\n".encode())
+        self.wfile.flush()
+        try:
+            while True:
+                try:
+                    msg = client.queue.get(timeout=30)
+                    self.wfile.write(msg.encode())
+                    self.wfile.flush()
+                except queue.Empty:
+                    self.wfile.write(": keepalive\n\n".encode())
+                    self.wfile.flush()
+        except (BrokenPipeError, ConnectionResetError):
+            pass
+        finally:
+            clients.pop(client.id, None)
+
+    def _handle_message(self):
+        params = parse_qs(urlparse(self.path).query)
+        client_id = params.get("client_id", [None])[0]
+        length = int(self.headers.get("Content-Length", 0))
+        body = json.loads(self.rfile.read(length)) if length else {}
+        response = self._process_jsonrpc(body)
+        if client_id and client_id in clients:
+            send_sse(client_id, "message", response)
+        self._json_response(response)
+
+    def _process_jsonrpc(self, msg):
+        method = msg.get("method", "")
+        msg_id = msg.get("id")
+        if method == "initialize":
+            return {"jsonrpc": "2.0", "id": msg_id, "result": {
+                "protocolVersion": "2024-11-05",
+                "capabilities": {"tools": {}},
+                "serverInfo": {"name": "handoff-pro", "version": "1.0.0"}
+            }}
+        elif method == "notifications/initialized":
+            return {"jsonrpc": "2.0", "id": msg_id, "result": None}
+        elif method == "tools/list":
+            tools_list = [{"name": k, "description": v["description"],
+                          "inputSchema": v["inputSchema"]} for k, v in TOOLS.items()]
+            return {"jsonrpc": "2.0", "id": msg_id, "result": {"tools": tools_list}}
+        elif method == "tools/call":
+            tool_name = msg.get("params", {}).get("name", "")
+            arguments = msg.get("params", {}).get("arguments", {})
+            if tool_name not in TOOLS:
+                return {"jsonrpc": "2.0", "id": msg_id, "error": {
+                    "code": -32601, "message": f"Unknown tool: {tool_name}"}}
+            try:
+                result = TOOLS[tool_name]["handler"](arguments)
+                return {"jsonrpc": "2.0", "id": msg_id, "result": {
+                    "content": [{"type": "text", "text": json.dumps(result, default=str)}]
+                }}
+            except Exception as e:
+                return {"jsonrpc": "2.0", "id": msg_id, "result": {
+                    "content": [{"type": "text", "text": json.dumps({"error": str(e)})}],
+                    "isError": True
+                }}
+        return {"jsonrpc": "2.0", "id": msg_id, "error": {
+            "code": -32601, "message": f"Unknown method: {method}"}}
+
+    def _json_response(self, data):
+        body = json.dumps(data).encode()
+        self.send_response(200)
+        self.send_header("Content-Type", "application/json")
+        self.send_header("Content-Length", len(body))
+        self.send_header("Access-Control-Allow-Origin", "*")
+        self.end_headers()
+        self.wfile.write(body)
+
+def main():
+    db.init_db()
+    server = HTTPServer(("0.0.0.0", PORT), MCPHandler)
+    print(f"Handoff Pro MCP server running on port {PORT}")
+    print(f"SSE endpoint: http://0.0.0.0:{PORT}/sse")
+    print(f"Tools registered: {len(TOOLS)}")
+    server.serve_forever()
+
+if __name__ == "__main__":
+    main()
diff --git a/tests/__init__.py b/tests/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/tests/test_workflows.py b/tests/test_workflows.py
new file mode 100644
index 0000000..86f43d5
--- /dev/null
+++ b/tests/test_workflows.py
@@ -0,0 +1,149 @@
+"""End-to-end workflow tests for Handoff Pro."""
+import json, os, sys, tempfile, unittest
+from unittest.mock import patch
+
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+
+from lib import db as db_mod
+from scripts import pricing_catalog, project_mgmt, estimate_generator
+from scripts import proposal_drafter, invoice_mgmt, daily_log
+from scripts import scope_generator, material_list, budget_tracker
+from scripts import project_scheduler, signature_mgmt, activity_logger, site_analyzer
+
+def fresh_db():
+    """Point to a fresh temp DB and initialize it."""
+    import tempfile as tf
+    path = tf.mktemp(suffix=".db")
+    db_mod.DB_PATH = path
+    db_mod.init_db()
+    return path
+
+class TestWorkflow1_NewEstimate(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+        pricing_catalog.seed_default()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_full_estimate_workflow(self):
+        # Generate estimate from description
+        result = estimate_generator.generate("kitchen remodel, 200sqft, new cabinets, countertops, flooring")
+        self.assertIn("id", result)
+        self.assertGreater(result["total_cost"], 0)
+        # Verify estimate retrievable
+        est = project_mgmt.get_estimate(result["id"])
+        self.assertEqual(est["id"], result["id"])
+        self.assertGreater(len(est["line_items"]), 0)
+
+class TestWorkflow2_Proposal(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+        pricing_catalog.seed_default()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_proposal_generation(self):
+        est = estimate_generator.generate("bathroom renovation, tile, plumbing, 100sqft")
+        result = proposal_drafter.draft(est["id"])
+        self.assertIn("proposal", result)
+        self.assertIn("Kellow Construction", result["proposal"])
+        self.assertIn("Payment Terms", result["proposal"])
+
+class TestWorkflow3_DailyLog(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_daily_log_workflow(self):
+        job = project_mgmt.create_job("Test Client", "123 Main St")
+        log = daily_log.create(job["id"], "Framed exterior walls", crew_names="Mike, Sarah", hours_worked=8)
+        self.assertIn("id", log)
+        logs = daily_log.list_logs(job["id"])
+        self.assertEqual(len(logs["logs"]), 1)
+
+class TestWorkflow4_Invoice(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+        pricing_catalog.seed_default()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_invoice_lifecycle(self):
+        est = estimate_generator.generate("deck build, 300sqft")
+        inv = invoice_mgmt.create(est["id"])
+        self.assertIn("invoice_number", inv)
+        self.assertGreater(inv["amount_due"], 0)
+        # Mark paid
+        paid = invoice_mgmt.mark_paid(inv["id"])
+        self.assertEqual(paid["status"], "paid")
+
+class TestWorkflow5_JobTreadSync(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    @patch("scripts.jobtread_sync.pave_query")
+    @patch("scripts.jobtread_sync.get_org_id", return_value="test_org")
+    def test_pull_jobs(self, mock_org, mock_pave):
+        mock_pave.return_value = {"organization": {"jobs": {"nodes": [
+            {"id": "jt123", "name": "Test Job", "status": "active", "createdAt": "2026-01-01",
+             "account": {"id": "acc1", "name": "John Smith"}, "location": {"address": "456 Oak Ave"}}
+        ]}}}
+        from scripts import jobtread_sync
+        result = jobtread_sync.pull_jobs()
+        self.assertEqual(result["synced"], 1)
+        # Verify in local DB
+        jobs = project_mgmt.list_jobs()
+        self.assertGreater(len(jobs["jobs"]), 0)
+
+class TestBudgetTracker(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+        pricing_catalog.seed_default()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_budget_report(self):
+        est = estimate_generator.generate("framing project, 500sqft")
+        job_id = project_mgmt.get_estimate(est["id"])["job_id"]
+        daily_log.create(job_id, "Day 1 framing", hours_worked=10)
+        report = budget_tracker.report(job_id)
+        self.assertIn("budget", report)
+        self.assertIn("actual", report)
+        self.assertEqual(report["actual"]["hours_worked"], 10)
+
+class TestSiteAnalyzer(unittest.TestCase):
+    def test_issue_detection(self):
+        result = site_analyzer.analyze("Found water damage in the corner near the window, also some cracking in foundation")
+        self.assertGreater(result["issues_found"], 0)
+        self.assertGreater(result["total_cost_impact"], 0)
+
+class TestScopeAndMaterials(unittest.TestCase):
+    def setUp(self):
+        self.db = fresh_db()
+        pricing_catalog.seed_default()
+
+    def tearDown(self):
+        os.unlink(self.db)
+
+    def test_scope_generation(self):
+        est = estimate_generator.generate("bathroom tile, plumbing fixtures")
+        scope = scope_generator.generate(est["id"])
+        self.assertIn("scope", scope)
+        self.assertIn("Scope of Work", scope["scope"])
+
+    def test_material_list(self):
+        est = estimate_generator.generate("kitchen cabinets, countertops, flooring")
+        mats = material_list.generate(est["id"])
+        self.assertGreater(mats["items_count"], 0)
+
+if __name__ == "__main__":
+    unittest.main()