- Home
- Skills
- Workflow Automation
- quantumclaw route optimizer
More screenshots
Works with the AI Agents you already use
quantumclaw route optimizer
Optimize delivery routes and benchmark classical vs. quantum-inspired solvers for VRP logistics.
Free
quantumclaw route optimizer
Example session with this skill installed
I have 3 vans with 10-unit capacity each and a CSV of 25 delivery stops in London. Optimize the routes to minimize CO2 and compare the classical solver against the quantum emulator.
- Read your context and instructions
- Compiled the quantumclaw route optimizer
I have processed the 25 delivery stops from your CSV and benchmarked the routes across the classical and dwave-sqa (quantum-inspired emulator) backends. The emulator produced the most efficient feasible plan.
Delivery Route Optimization Report
Benchmark Seed: 1
Winner: dwave-sqa (Quantum-inspired Emulator)
Performance Summary
| Metric | Baseline | Optimized (dwave-sqa) | Improvement |
|---|---|---|---|
| Total Distance | 84.2 km | 68.5 km | -18.6% |
| Total Cost | £142.10 | £118.40 | -16.7% |
| CO2 Emissions | 22.4 kg | 18.2 kg | -18.7% |
| Lateness | 12 min | 0 min | -100% |
Optimized Fleet Schedule
- Van-1 (8/10 units): Depot -> Stop #4 -> Stop #12 -> Stop #3 -> Stop #19 -> Stop #2 -> Depot
- Van-2 (9/10 units): Depot -> Stop #7 -> Stop #1 -> Stop #15 -> Stop #22 -> Stop #10 -> Stop #5 -> Depot
- Van-3 (8/10 units): Depot -> Stop #9 -> Stop #11 -> Stop #6 -> Stop #25 -> Stop #14 -> Depot
Note: The classical solver was 4.2% less efficient in total distance and failed to assign Stop #14 in 2 out of 5 runs due to tight capacity constraints. All time windows in the winning plan are met.
Next steps
- Review the specific stop sequences in
plan.jsonfor driver dispatch. - Provide a road-distance matrix if you need to account for specific traffic routing or one-way streets.
- Run with
--repeat 20if you want to further narrow the median cost variance.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing last-mile delivery logistics is a high-stakes balancing act where inefficient routes waste fuel, time, and money. This skill implements the QuantumClaw Q-Router to solve Vehicle Routing Problems (VRP) with capacity constraints and time windows, allowing you to optimize fleets and benchmark different solver technologies side by side.
What it does
- Problem validation parses stops, vehicles, and depots from CSV or JSON to ensure routing models are mathematically sound before execution.
- Multi-solver benchmarking runs classical heuristics, simulated annealing, and quantum-inspired emulators against the same problem set.
- Performance ranking evaluates plans based on total service completion first, then optimizes for cost, distance, and CO2 emissions.
- Comparative analysis generates reports that contrast optimized results against your baseline routes to quantify actual savings.
- Fleet assignment produces specific stop-by-stop schedules for each vehicle in the fleet.
How it works
- Model building converts your logistics data into a validated Q-Router problem file, incorporating vehicle capacities and delivery windows.
- Solver execution triggers the
quantumclawCLI to run parallel benchmarks across multiple backends includingclassicalanddwave-sa. - Plan selection extracts the cheapest feasible plan from the benchmark results, ensuring 100% of deliveries are served.
- Reporting generates a Markdown summary of routes, KPIs, and required human interventions for late arrivals or capacity issues.
Frameworks & tools
This skill utilizes the QuantumClaw Rust-based CLI, Python 3.9+, and the D-Wave Ocean SDK for simulated and quantum-inspired annealing. It operates entirely offline for local solvers.
Why this beats prompting it yourself
Generic LLMs struggle with the combinatorial complexity of VRP and often "hallucinate" feasible routes that actually violate vehicle capacities or time windows. This skill uses the LLM as a controller for specialized optimization engines, ensuring every route is mathematically validated and reproducible via specific seeds.
Use cases
- Last-mile optimization to reduce daily mileage and fuel costs for delivery fleets.
- Solver benchmarking to compare classical heuristics against quantum-inspired annealing for complex logistics.
- Logistics auditing to identify inefficiencies in existing manual route assignments.
- Capacity planning to determine if a fleet can handle increased demand without adding vehicles.
Known limitations
Distances default to haversine estimates rather than road-network paths unless an explicit distance matrix is provided. Real quantum hardware access requires an external D-Wave API token.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.
Distances use straight-line haversine estimates by default. Real quantum hardware requires a D-Wave API token. The classical solver may leave stops unassigned if fleet capacity is near 100%.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 23 days ago
- Free with an Agensi account
- Software
Python