Borrowing it
Nothing to install: this file belongs to TianLin0509/SuperRAN. 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/TianLin0509/SuperRAN/main/CLAUDE.mdgit clone --depth 1 https://github.com/TianLin0509/SuperRANWrote 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/instructions/tianlin0509/superran/claude-md)<a href="https://agentmods.dev/instructions/tianlin0509/superran/claude-md"><img src="https://agentmods.dev/badge/instructions/tianlin0509/superran/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.37402 | $0.37402 |
| Opus 5 | $0.18701 | $0.18701 |
| Sonnet 5 | $0.07480 | $0.07480 |
| Haiku 4.5 | $0.03740 | $0.03740 |
Grade A, and why
SuperRAN CLAUDE.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 today.
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 — 1,766 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SuperRAN 开发规范
Agent 协作入口（先读这里）
SuperRAN 由一位维护者（无线通信工程师）主导，Agent 是执行者不是决策者。
所有协作规则在 .agents/ 目录，开工前必读，不要按聊天里粘贴的 prompt 工作：
- 实现任务：
.agents/AUTHOR.md - 独立审核：
.agents/MERGER.md - 需要几个 Reviewer：
.agents/RISK.md（按文件路径查表，不许自己估） - 推送与建 PR：
.agents/SYNC.md（只在维护者明确说“同步 GitHub”时执行） - 仿真设计、数据生成或性能结论：
skills/channel-sim/SKILL.md
三条铁律：一个提交只动一个物理机制；审核发现的物理 bug，修复时必须补一条
“revert 掉就会变红”的测试并入 tests/test_physics_invariants.py；不许静默降级，
用了工程近似要写进报告的“没证明什么”。
报告统一用 python scripts/make_agent_report.py <report.json> 生成，字段照抄
.agents/report.example.json，不要手写 HTML。看当前状态用
python scripts/superran_board.py。
develop 是唯一主线，主工作区就是 C:\Vibe\Wireless\SuperRAN；并行任务用
git worktree add 建到 C:\Vibe\Worktrees\SuperRAN\<任务>，禁止再 clone 一份。
Agent 默认不 push、不建 PR、不合并远端。
skills/superran-lead/与skills/superran-member-task/是已废弃的多人「组长-组员」 流程（含 FORMAL/REHEARSAL 模式、Fork 推送等），不要再按它们工作，一律以.agents/为准。
与人交流采用“双层表达”：保留准确技术术语，同时紧跟一句白话解释；关键物理概念再给 一个贴近当前任务的小例子。例子只帮助理解，不能冒充代码事实、测试证据或性能结论。 首次使用缩写时展开全称；实现前用“我理解为……，不等于……”复述边界。专家已明确 理解时不要反复教学。
项目定位
SuperRAN 是独立维护、独立演进的 Agent 式无线仿真平台。信道轴序、
TDD 互易、预编码、功率约束、RBG/TBS、链路自适应和 KPI 都以
SuperRAN 本仓的合同为准。统计信道生成、CDL/TDL 表、NR 载波/TDD、参考序列、
阵列、LMMSE 与几何 S/N/I 均由 src/superran first-party 实现；运行时不得
搜索、导入或修改 MSG-Platform / ChannelHub 源码树。
channelhub.py 只保留为历史 Python API 的兼容门面，实际实现固定指向
native.py。历史环境变量 SUPERRAN_CHANNELHUB 被有意忽略，不能改变生成字节。
Sionna RT 只能作为显式 direct optional adapter 接入；不可用时硬报告，
不得退回另一个仓库。QuaDRiGa 路线已于 2026-09-04 明确不做并从代码与文档中删除
（需要 MATLAB/Octave 运行时，成本与收益不成比例）；要空间一致性就按 38.901 §7.6.3
自己实现一个子集。
默认信道是 CDL（internal_sim）。sionna_rt 是本仓自己的直连适配层
（src/superran/sionna_rt.py），装了 sionna-rt 才可用，必须在配置里显式写
source: sionna_rt 才会走。引擎清单恒为这两条。
环境
- Python ≥ 3.10，需要 numpy / scipy / pydantic v2 / pyyaml / structlog / mcp
- 射线追踪需
pip install sionna-rt（连带 mitsuba + drjit，约 300 MB）； direct adapter 已实现，但本地 OSM 资产、材料标定与多端口垂直相位仍须按报告边界验证
测试
python tests/test_e2e.py # 端到端
python tests/test_mcp_server.py # MCP 全链路
python tests/test_raytracing.py # 射线追踪与决策层
python tests/test_linklevel.py # 谱效、可信度、物理层、IRC
python tests/test_gates.py # 校准、标准表、三道门、统计判决
python tests/test_results.py # 外部算法结果契约、预注册
python tests/test_linkadapt.py # 链路自适应、吞吐、并行生成
python tests/test_mumimo.py # MU-MIMO、单码字、RBG 粒度
python tests/test_system.py # 系统级仿真（单路径；容量=full_buffer 话务）
python tests/test_scheduler_p0.py # 调度资源账本、频选、MU 评分与 Finalizer
python tests/test_srs_resource.py # PCI 模3、4 CS、2T4R 双腿与全局周期容量
python tests/test_srs_waveform.py # RE级波形、解扩、CFO/时偏与UL IoT证据
python tests/test_interference.py # IoT、预设、说明书、算法页、文档计数
python tests/test_csi_aging.py # CSI 时延、SRS 跳频、真实/估计视角
python tests/test_rng.py # 随机数分流、重复实验、CRN、置信区间
python tests/test_sysscenes.py # 系统场景预设与成对受控性
python tests/test_power_control.py # EBF/PEBF/NEBF 与逐 RB 功率耦合
python tests/test_physics_contract_extensions.py # 快照时钟、SRS测量口径与场景资产合同
python tests/test_physics_invariants.py # 极化、子阵、SRS/LMMSE 物理不变量
python tests/test_channel_generation_contract.py # first-party 信道生成合同与最小网格
python tests/test_native_independence.py # 外部根/导入阻断、v1/v2互易、35工具
python tests/test_developer_guide.py # 开发者文档覆盖、离线结构与漂移检查
python tests/test_carrier.py # 载波栅格、Type-0 边界、本地 TDD 合同
python tests/test_company_256t.py # 256T 阵列与码本
python tests/test_system_sim_tool.py # sr_system_sim 行为级（硬失败路径）
python tests/test_benchmarks.py # 预注册经典通信基准与 provenance
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.
- today Changed · +331 lines · +7,906 tokens per session d91e1e3fb821
- 2d ago Changed · +8 lines · +206 tokens per session 0bb627715198
- 3d ago Changed · +970 lines · +20,911 tokens per session df99d0843a57
- 7d ago First seen · 457 lines · 8,379 tokens per session scan A a4bdcb56b8cd
SuperRAN CLAUDE.md is an instructions file published in the GitHub repository TianLin0509/SuperRAN (2 stars, last pushed yesterday), licensed MIT. It adds 37,402 tokens to every session, about $0.1870 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 2026-08-31.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.