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.
git clone --depth 1 https://github.com/FuRongJun-1999/dsh-memoryWrote 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/rules/furongjun-1999/dsh-memory/rule)<a href="https://agentmods.dev/rules/furongjun-1999/dsh-memory/rule"><img src="https://agentmods.dev/badge/rules/furongjun-1999/dsh-memory/rule/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/rules/furongjun-1999/dsh-memory/rule"><img src="https://agentmods.dev/badge/rules/furongjun-1999/dsh-memory/rule.svg" alt="Reviewed on agentmods" width="80" 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 | $0.07898 | $0.07898 |
| Opus 5.5 | $0.03159 | $0.03159 |
| Sonnet 5.5 | $0.01580 | $0.01580 |
| Haiku 4.5 | $0.00790 | $0.00790 |
Grade A, and why
RULE scanned grade A with 1 finding 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 9d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **动作**：构造 argv 列表→subprocess.run(capture_output=True, text=True, encoding='utf-8', errors='replace')→env 带 PYTHONUTF8=1→shell=False；边界判据=本次调用是否产出非UTF-8字节流(纯MSYS工具与不起进程的内置工具豁免；原生Windows程序不豁免，须python包装或显式解码)；仓内引用与文档路径一律写 How it starts
The opening of the file, as written. The whole thing — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
工作纪律（工作纪律_认知图条目_v1.1）· 18 条
本文件由
scripts/render_discipline.py从真源docs/工作纪律_认知图条目_v1.1.json渲染生成，请勿手改；改真源后重跑渲染。 真源指纹（SHA256 前16位）：9486668b0dadac39
概述
工作纪律集: 4条方法论(理论先行/全面处理/白箱方法/根因) + 3条执行(验证/双副本/兜底) + 1条思考(中文思考/协作模式分级) + 1条保密(敏感信息隔离) + 3条图像线(选源护栏/历史查询优先/算法权威唯一) + 1条协作(访谈澄清:先问清再动手) + 1条合规(内容政策合规:公开产物过内容政策+隐私双清单) + 2条执行调度与运行环境(命令执行统一走python/UTF-8: 显式encoding+PYTHONUTF8=1, 规避GBK, 全仓文本与路径一律UTF-8(无BOM, 中文可进路径), 不做控制台兼容(取用走python), 公开面路径写相对路径(守 check_local_paths); 任务派发统一走蜂巢(zcode端的工作流工具为等效蜂巢工具——使用者2026-10-04裁定; 蜂巢功能完全完善之前可优先使用zcode工作流): 任何执行性任务经蜂巢spawn/submit执行并留痕, agent本体只做编排, L1出口=产物落点为只读判定(不改仓库/外部状态)可直跑须留痕) + 1条记忆闭环(任务收尾归档:每次任务执行完只将核心修改(内容/原因/位置/验证结论)按CCG六要素(功能名/生效条件/子功能/执行/验证方式/不适用条件)成文存入灵枢记忆, 不写中间过程; text 类缺要素即被写入闸门拒并当场返回缺失清单) + 1条信息获取(工作区索引优先: 查工作区文件先读 WORKSPACE_INDEX.md(仓根, 由 scripts/workspace_index.py 从 git 追踪面管线生成), 尚无则先跑 --write 生成再读, 不以重复全盘浏览代替; 索引是会陈化的手工件, 故须管线生成+机械守卫, 否则退化为误导源)
执行公约
| 约定 | 内容 |
|---|---|
| 0.1 会话起始先查记忆 | 动手前先调认知图入口——工具名随端：MCP 端 = server「mdcg」的 cg；DSH 端 = 内建基元 lingshu_cg（即下文所有 cg(...)，同一服务勿因名字不同而猜测）。op 是必填参数（route/read/write/verify/review…），任何调用都显式传、勿省略。先 cg(op=route, intent=<任务意图>)；命中纪律/知识则按其 execution.how 执行。禁止跳过此步直接改代码或跑命令。 |
| 0.2 命中即声明 | 命中纪律须在回复中输出该条 response.direct 原文（「按工作纪律第 N 条：…」）。这是「已执行」的唯一可观测证据；未声明即视为未执行。 |
| 0.3 产出即证据 | 「动作」栏要求的产出（判定单／检索结果／验证结论／归档节点 id）必须实际给出，缺失即未执行。 |
| 0.4 违规即归档 | 发现自己违反任一条 → 第 4 条归因 + 第 16 条写入灵枢记忆。隐瞒违规本身是更严重的违规。 |
| 0.5 例外不外溢 | 某条的「不适用」只豁免该条，不豁免其它条。 |
纪律路由表
1. 理论先行
- 触发：重要项目｜长期任务
- 动作：识别项目级别→检索理论文档与既有实践→输出理论要点→再进入设计
- 不适用：情感交互｜闲聊
- 声明：按工作纪律第1条: 理论先行——先查相关理论与既有实践再动手。
2. 全面处理
- 触发：有相关记忆｜有认知图｜有对应权限
- 动作：读任务要求→读工作记忆→查相关任务→(有成果:复用 | 无:设计+调研)
- 不适用：情感交互｜闲聊
- 声明：按工作纪律第2条: 全面处理——先读任务/工作记忆, 有相关成果就在其上开发, 没有则先设计调研。
3. 白箱方法
- 触发：trained_on_4docs_intro
- 动作：识别任务条件→找知识/规则(并行递归)→精准执行; 正确→记录; 错误→找条件
- 不适用：需要快速执行的短期事项｜情感交互｜闲聊
- 声明：按工作纪律第3条: 白箱方法——用认知图【识别条件→找知识规则→精准执行】, 正确记录/错误找条件/不猜测/未验证不写入。
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.
- 9d ago Changed · +123 tokens per session ecb128383e4d
- 10d ago Changed · +158 tokens per session 5210ef40152c
- 11d ago Changed · +195 tokens per session d202360e6e29
- 12d ago Changed · +2 tokens per session a5d99fb8ebac
- 16d ago Changed · +117 tokens per session 0f2bdd6bedcc
- 17d ago Changed · +539 tokens per session deb9f71ada01
- 22d ago First seen · 212 lines · 6,764 tokens per session scan A ca732f79e2f5
RULE is a cursor rule published in the GitHub repository FuRongJun-1999/dsh-memory (396 stars, last pushed today), licensed MIT. It adds 7,898 tokens to every session, about $0.0316 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-24.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
shared-plane-packages
Pattern for creating shared plane packages that serve both migration and runtime planes.