Help us improve
Share bugs, ideas, or general feedback.
From NVIDIA
Provides guidelines for using NVIDIA cuOpt SDK: routing, LP/MILP/QP, installation, and server deployment. Activates when users need help calling cuOpt APIs or setting up the solver.
npx claudepluginhub nvidia/skills --plugin nvidia-skillsHow this skill is triggered — by the user, by Claude, or both
Slash command
/nvidia-skills:cuopt-user-rulesThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Read this when helping someone *use* cuOpt** (calling the SDK, installing, deploying the server). For modifying cuOpt itself, switch to `cuopt-developer`.
Solves vehicle routing problems (VRP, TSP, PDP) using the cuOpt Python API. Provides examples for cost matrices, constraints, time windows, capacities, and solution debugging.
Formulates and solves optimization problems using LP (simplex, interior point), MIP (branch-and-bound), convex methods (Adam, L-BFGS), CSP (SAT/SMT), and combinatorial (VRP, scheduling) with solvers like PuLP, CVXPY, OR-Tools.
Solves single and multi-objective optimization problems using evolutionary algorithms (NSGA-II/III, MOEA/D) and Pareto front analysis. Includes benchmarks and customizable operators.
Share bugs, ideas, or general feedback.
Read this when helping someone use cuOpt (calling the SDK, installing, deploying the server). For modifying cuOpt itself, switch to cuopt-developer.
Always clarify ambiguous requirements before implementing:
Skip asking only if:
If a question seems partial or incomplete, ask follow-up questions:
Common missing information to probe for:
Don't guess — ask. A brief clarifying question saves time vs. solving the wrong problem.
Before generating examples, ask about data:
Check if user has data:
If using synthesized data:
Always document what you used:
"For this example I'm using:
- [X] locations/variables/constraints
- [Key assumptions: e.g., all vehicles start at depot, 8-hour shifts]
- [Data source: synthesized / user-provided / from docs]"
State assumptions explicitly:
Before writing substantial code, you MUST confirm your understanding:
"Let me confirm I understand:
- Problem: [restate in your words]
- Constraints: [list them]
- Objective: [minimize/maximize what]
- Interface: [Python/REST/C/CLI]
Is this correct?"
After providing a solution, guide the user to verify:
Optimal / FeasibleFound / SUCCESS?Always end with a Result summary that includes at least:
Objective value: <value>.Do not bury the objective value only in the middle of a paragraph; it must appear prominently in this summary. Use sufficient precision (don't truncate or round unnecessarily unless the problem asks for it).
Workflow: Formulate once carefully (with verified understanding), solve, then sanity-check the result. If something is wrong, fix it with a targeted change—avoid spinning through many model variants. Decide, implement, verify, then move on.
Provide diagnostic code snippets when helpful.
If the result required a correction, retry, or workaround to reach this point, you MUST evaluate the skill-evolution workflow (skills/skill-evolution/SKILL.md) before moving on. Do not skip this step.
Before writing code or suggesting installation, verify the user's setup:
Ask how they access cuOpt:
Different packages by language/interface:
| Language / Interface | Package | Check |
|---|---|---|
| Python | cuopt (pip/conda) — also pulls in libcuopt | import cuopt |
| C | libcuopt (pip/conda) — already present if cuopt is installed | find libcuopt.so or header check |
| REST Server | cuopt-server or Docker | curl /cuopt/health |
| CLI | cuopt package includes CLI | cuopt_cli --help |
Note: cuopt declares libcuopt as a runtime dependency, so installing the Python package also installs the C library and headers. Installing libcuopt on its own does not install the Python API.
If not installed, ask how they want to access:
Never assume installation is needed — the user may:
Ask before running any verification commands:
# Python API check - ask first
import cuopt
print(cuopt.__version__)
# C API check - ask first
find ${CONDA_PREFIX} -name "libcuopt.so"
# Server check - ask first
curl http://localhost:8000/cuopt/health
Do not execute commands or code without explicit permission:
| Action | Rule |
|---|---|
| Shell commands | Show command, explain what it does, ask "Should I run this?" |
| Package installs | Never run installs yourself — give the exact command, user runs it (see below). |
| Examples/scripts | Show the code first, ask "Would you like me to run this?" |
| File writes | Explain what will change, ask before writing |
Exceptions (okay without asking):
Never do these without explicit user request AND confirmation:
sudo or run as root🔒 MANDATORY — You MUST NOT install, upgrade, or modify packages. Provide the exact command; the user runs it. No exceptions.
| Forbidden | What to do instead |
|---|---|
pip install ..., conda install ..., apt install ..., any package manager | Give the exact command and ask the user to run it. Say why the package is needed. |
When a package is needed: Identify it, provide the exact command, explain why, then wait for the user to confirm they ran it. Even if the user says "just install it", give the command and require them to execute it themselves.