Borrowing it
Nothing to install: this file belongs to xi-zhao/OpenQuantum. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/xi-zhao/OpenQuantum/main/.agents/skills/qpanda-qubo/SKILL.mdgit clone --depth 1 https://github.com/xi-zhao/OpenQuantumWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xi-zhao/openquantum/qpanda-qubo)<a href="https://agentmods.dev/skills/xi-zhao/openquantum/qpanda-qubo"><img src="https://agentmods.dev/badge/skills/xi-zhao/openquantum/qpanda-qubo/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xi-zhao/openquantum/qpanda-qubo"><img src="https://agentmods.dev/badge/skills/xi-zhao/openquantum/qpanda-qubo.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00128 | $0.01729 |
| Opus 5.5 | $0.00051 | $0.00692 |
| Sonnet 5.5 | $0.00026 | $0.00346 |
| Haiku 4.5 | $0.00013 | $0.00173 |
Grade A, and why
qpanda-qubo 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 15d ago.
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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QPanda QUBO Workbench
原生 SDK 路径
本能力同时支持 QPanda Algorithm 的原生公开 API。用户需要 QUBO 建模、约束、算法与上游优化模块 时,由 Agent 在固定依赖环境中编写和运行任务代码,不限于下文便捷 Tool 的预设模型。
使用 node scripts/sdk-access.mjs inspect qpanda-qubo 定位环境,按完整 SDK 工作流执行。模块导航:pyqpanda_alg;导航不是 API 白名单。
macOS 的完整 SDK setup 需要 Git、CMake 和 C++ 编译器,以固定 fmt 11.2.0 源码构建隔离动态库;其余上游 wheel 的系统依赖仍按平台核验,不修改 Homebrew 或裁剪算法模块。
原生程序导入库前,按环境定位器的 bootstrap 执行启动兼容处理:macOS 从独立目录加载固定 fmt 11 ABI,保留 pyqpanda_alg 顶层完整导入。这不限制公开 API,也不替换数值算法。
下文关于模型、门集、输入与预算的限制描述便捷 Tool 的合同;原生 SDK 另受上游版本、可选依赖、宿主资源与用户授权约束。已禁用或被拒绝的动作不能借代码路径绕过。执行状态和科研验收分别报告。
工作边界
这个 Skill 负责组织 QUBO 求解工作流,真正的计算由 qpanda_qubo MCP Server 暴露并经 Harness MCP Client 注册的 Tool 完成,
底层调用本源官方 pyqpanda_alg 的 QUBO 模块(QuadraticBinary / QUBO_QAOA)。当前只开放
本地、无凭据的求解;调用前须显式准备固定 Python 环境(pyqpanda3 是原生 wheel):
- 二值变量数由模型决定;
- 可以用变量名、minimize/maximize 目标和线性等式约束建模;每个约束必须显式给出 penalty;
- QUBO 以数值系数给出:
quadratic(方阵)、可选linear、可选constant; method=traversal始终返回经典暴力遍历的确定性最优解;method=qaoa运行本地 QAOA(layer由用户选择),返回比特串概率分布。
当前不开放不等式自动松弛、penalty 自动选择、本源量子云、Token、真机任务提交、任意 Python、 sympy 表达式字符串、文件路径或数据集加载。
工作流
- 按用户输入选择下述主动作;主动作会加载固定环境并随结果返回
packageVersion,不需要额外的运行时检查调用。 - 若 Tool 不存在,告诉用户在“设置中心 → 量子组件 → MCP Server 连接”把 QPanda QUBO 配置为启用,重启 OpenQuantum 后再试;不要改用 Bash 绕过设置。
- 用户给的是业务目标和等式约束时,优先调用
model_and_solve_qpanda_qubo。检查constraints.feasible和penalty.sufficient;未检查时不得宣称可行性或 penalty 已验证;后者失败表示 penalty 太弱,不能把 QUBO 最优解写成 原约束问题的最优解。 - 用户已经给出 QUBO 数值时,调用
solve_qpanda_qubo:目标是x^T Q x + b·x + c,quadratic是Q,linear是b,constant是c。 - 按任务选择 traversal 或 qaoa;QAOA 不要求先遍历。两项 Tool 都接受 referenceMode;auto 仅在变量数不超过 12 时穷举,required 尝试所请求规模,skip 不做额外穷举。
- 比较 QAOA 概率分布的最高比特串与经典最优解时,注意比特序可能不同,按目标函数值判断而不是 直接按比特位比较。
- 输出时区分:用户模型、编译后的 QUBO、上游工具事实、枚举 observations、尚未完成的来源链验收。
与其他能力的分工
- 电路层面的构建、转译与文档查询用
qiskit-circuit-workbench或tyxonq-workbench。 - 严格限定的二量子位基态 VQE 与独立验收用
quantum-ground-state。 - 真机执行(悟空 QPU)使用默认关闭的
qpanda_runtimeMCP Server 所暴露的 Tool;本地 QUBO 求解不提交任何云任务。
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 15d ago Changed · +15 lines a38009500dbe
- 1mo ago Changed · +4 lines 2058f9a3b9e8
- 1mo ago Changed · +2 lines · -4 tokens per session 9565a823f9d9
- 1mo ago Changed f753f22cd50d
- 1mo ago First seen · 64 lines · 132 tokens per session scan A 1321094a52dd
qpanda-qubo is a skill published in the GitHub repository xi-zhao/OpenQuantum (71 stars, last pushed today), licensed MIT. It adds 128 tokens to every session and 1,729 once invoked, about $0.0005 per session on Opus 5.5. 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 skills, from other repositories
cancel-or-change-award
Works out the cancel or change fee, where the points go back, and if a rebook is cheaper than keeping the ticket. Use when the user wants to cancel, change or rebook an award ticket. Requires the Award Travel Finder MCP connector (https://mcp.awardtravelfinder.com/mcp).
maximize-transfer-bonus
Decides if a live transfer bonus is worth using now, what to book with it and how many points to move. Use when the user asks about a transfer bonus, or a bonus to one of their programmes is live. Requires the Award Travel Finder MCP connector (https://mcp.awardtravelfinder.com/mcp).
x402-discovery
Agents discover paid artefact rails via GET https://councilof.ai/.well-known/x402.json (mode is reported from runtime env, Base eip155:8453, USDC; amounts from challenge only — no invent price). Catalog GET /api/x402. Free door GET /api/free-door amount 0 settles and charges nothing — proves the rail. Living board GET…
sherweb
Every Sherweb Partner API capability, plus a local SQLite store, offline analytics, and margin/drift/orphan joins no other Sherweb tool has. Trigger phrases: reconcile sherweb billing, what's my margin per customer in sherweb, find orphaned sherweb subscriptions, list sherweb customers, preview a sherweb seat change…
business-model
Create or review business model deliverables with traceable assumptions, decision criteria, ownership, and risk controls. Use when the user asks for business model planning or analysis.
emil-design-eng
This skill encodes Emil Kowalski's philosophy on UI polish, component design, animation decisions, and the invisible details that make software feel great.