Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/nvidia/cuopt/agents-mdgit clone --depth 1 https://github.com/NVIDIA/cuoptWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00913 | $0.00913 |
| Opus 5 | $0.00456 | $0.00456 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00091 | $0.00091 |
Grade A, and why
cuopt AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — cuOpt AI Agent Entry Point
AI agent skills for NVIDIA cuOpt optimization engine. Skills live in skills/ (repo root) and use a flat layout: per domain, a concept skill (formulation / problem types) plus implementation skills — typically one per interface (Python, C, CLI, server), or consolidated when the content is shared across interfaces (e.g. installation). Skills evolve through agent interactions — see skills/cuopt-skill-evolution/ for the evolution workflow.
🔒 MANDATORY — Ambiguity: When the problem could be read more than one way, you MUST either ask the user to clarify or solve every plausible interpretation and report all outcomes. Never pick one interpretation silently.
Skills directory (flat)
Rules
skills/cuopt-user-rules/— Base rules for end users calling cuOpt (routing, LP, MILP, QP, install, server). Not for cuOpt internals — seeskills/cuopt-developer/. Read first for user-facing tasks; choose skills from the index below by task and interface.skills/cuopt-developer/— Modify, build, test, debug, and contribute to cuOpt internals (C++/CUDA, Python, server, CI). Use for solver internals, PRs, DCO, and code conventions. For VRP dimension work (combine invariants, fwd/bwd propagation, new constraints/objectives in the routing engine), readskills/cuopt-developer/references/vrp_skills.mdin addition toSKILL.md.skills/cuopt-skill-evolution/— Skill evolution: after solving a non-trivial problem, propose skill updates to capture generalizable learnings.
Common (concepts only; no API code)
skills/cuopt-numerical-optimization-formulation/— LP / MILP / QP: concepts + problem parsing + common formulation patternsskills/cuopt-multi-objective-exploration/— Multi-objective: trace + interpret the Pareto frontier across competing objectives (ε-constraint / weighted-sum over repeated cuOpt solves)
Installation
skills/cuopt-install/— User install for Python, C, and server (pip, conda, Docker, verification). For building cuOpt from source, seeskills/cuopt-developer/.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 54 lines · 913 tokens per session scan A beb5c6e963c3
cuopt AGENTS.md is an instructions file published in the GitHub repository NVIDIA/cuopt (1,032 stars, last pushed 3d ago), licensed Apache-2.0. It adds 913 tokens to every session, about $0.0046 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
flashinfer AGENTS.md
Instructions for flashinfer-ai/flashinfer: This repository’s agent instructions live in CLAUDE.md.
cuda-metal AGENTS.md
Instructions for Lulzx/cuda-metal, covering agents.md — cumetal development guide, project identity, current focus, repository structure and non-negotiable constraints.
VeloQ AGENTS.md
Instructions for lucifer1004/VeloQ, covering veloq — contributor guidelines, wire-format invariants (do not break casually), workspace layout, shipped commands (status roadmap) and code conventions.
gpu-server-setup AGENTS.md
Instructions for EvilFreelancer/gpu-server-setup, covering authoring & maintaining the gpu-server-setup skill, repository layout, editing rules, keeping metadata in sync and what this skill does not do.
AutoMegaKernel AGENTS.md
Instructions for RightNow-AI/AutoMegaKernel, covering what amk is, the edit surface, scheduleconfig + kernelknobs only, canonical tool surface (mcp), use these exact names, codex mcp server config (/.codex/config.toml) and the copy-paste loop a codex agent runs.
flashinfer CLAUDE.md
Instructions for flashinfer-ai/flashinfer, covering claude.md, project overview, quick reference, quick start for development and installation.