Borrowing it
Nothing to install: this file belongs to lllyin/mcp-cn-a-stock. 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/lllyin/mcp-cn-a-stock/main/AGENTS.mdgit clone --depth 1 https://github.com/lllyin/mcp-cn-a-stockWrote 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/lllyin/mcp-cn-a-stock/agents-md)<a href="https://agentmods.dev/instructions/lllyin/mcp-cn-a-stock/agents-md"><img src="https://agentmods.dev/badge/instructions/lllyin/mcp-cn-a-stock/agents-md/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/instructions/lllyin/mcp-cn-a-stock/agents-md"><img src="https://agentmods.dev/badge/instructions/lllyin/mcp-cn-a-stock/agents-md.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 | $0.01252 | $0.01252 |
| Opus 5.5 | $0.00501 | $0.00501 |
| Sonnet 5.5 | $0.00250 | $0.00250 |
| Haiku 4.5 | $0.00125 | $0.00125 |
Grade A, and why
mcp-cn-a-stock 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 1mo 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.
What it actually says
项目开发约束
这是一个开源的 A 股数据服务。
运行与发布
- 服务启动、停止和重启统一通过
start.sh/stop.sh,不另起进程或手工改 pidfile。 - 发布前必须在待发布版本上运行
scripts/verify_release.py并通过验证,保留报告。
一、优先级:数据完整 > 功能正确 > 性能
- 任何改动只要可能让某一维数据变少、变旧、变不准,就不做——哪怕它更快、更省内存。 缺一段数据用户看得见,快 200 毫秒看不见。
- 不为省资源换一条数据更差的路。
- 结论是"不值得改"同样是有效产出,写进注释即可。
二、取数:每一维都要能换源
- 一维数据不能只挂在一个源上。 加一个源的成本要低到愿意加——做不到是抽象没提对, 不是源难接。
- 回退是常态不是异常。 上一个源没给出结果就问下一个,空结果同样往下走。
- 每一级失败留下能定位到源的日志。 查一个数是谁给的,不该靠猜。
- 各给一部分字段就合起来用。 对同一字段给出不同的值是仲裁问题:按固定顺序判, 把差异记进已知差异清单——不能让"谁先返回"决定数据。
落地形态见 docs/architecture.md,两者冲突时以上面四条为准。
三、性能:有界、可比、等价
- 目标是降低 MCP 调用耗时,尤其多标的多批次并发下的 P95 和最长耗时。
- 消除排队、重复请求、串行等待。线程、任务、队列、连接池、缓存一律有界且可回收。 任何「N × 单次超时」的结构都要有总预算——调小单次超时只是换个系数。
- 验证覆盖单批 4 个标的,以及同时 5 批、10 批,比端到端耗时、P90/P95、最大、吞吐、错误率。
- 优化不得改变返回数据。前后用同一组请求做输出比对(逐字或哈希)证明等价:覆盖钉日期 和实时两类,先关掉报告缓存,并先用同一份代码连跑两次确认基线稳定——上游本身会抖。
四、资源与环境
- 峰值内存 ≤ 500 MiB,按服务及其全部子进程(浏览器、Xvfb 这些)的合计峰值算, 超限的性能提升一律不采纳。
- Ubuntu 优先,不依赖 macOS 专属能力。发布前在目标环境验证启动、停止、并发、 浏览器回退和内存峰值。
五、改动前先算,值要量出来
- 改关键配置、加缓存/回退/熔断、调并发上限,先写下三件事:省的是哪个可观测量、省多少; 代价落在哪(同一条链路上还有哪些阈值会跟着变);碰不碰数据完整性。有一件写不出来 就先别写代码。
- 可观测量指:命中率、P95、吞吐、峰值内存、上游请求数、积分消耗、数据陈旧度。 不能只说"更快"或"更稳"。收益说不清量级的、与副作用相当的,保持默认不动。
- 优先用生产日志或归档回放估收益,不靠推理。 已有机制可能已经把收益吃掉了。
- 参数的值要量出来,不是拍出来。 能自动量的做成
scripts/probe_tuning.py的一项, 让它出推荐值和证据。预算/上限类参数按最大值定,不按分位数——分位数按定义会砍尾部, 而那正是要保住的部分。 - 报"没问题"之前,先证明这个测量测得出问题(注入一次已知故障看它报不报)。
六、这是开源项目,不绑任何一台机器
- 配置项的注释只写两样:这个参数是什么含义、怎么为自己的环境取值。不写某次事故的 时间、请求 ID、逐条实测秒数——那些归 commit message 和 git 历史。
- 措辞用「当前环境」,别写「本机」「部署机」这类指某一台机器的说法。
七、提交信息精简、不带署名
- 禁止任何署名或协作者 trailer(
Co-Authored-By、Generated with、Signed-off-by)。 - 一行标题说清改了什么;正文最多三五行,只留判断依据。不贴时间线、不复述代码。 只有 breaking change 才允许写长,写明破坏了什么、调用方怎么改。
- 详细推导写进代码注释或文档——那里才会被下一个人读到。
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.
- 1mo ago Changed · +45 lines · +912 tokens per session 3aa5b336bdda
- 2mo ago First seen · 24 lines · 340 tokens per session scan A b778a9364079
mcp-cn-a-stock AGENTS.md is an instructions file published in the GitHub repository lllyin/mcp-cn-a-stock (5 stars, last pushed 1mo ago), licensed MIT. It adds 1,252 tokens to every session, about $0.0050 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 2026-08-31.
mcp-cn-a-stock AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| mcp-cn-a-stock AGENTS.mdlllyin/mcp-cn-a-stock | Instructions file | 5 | 1mo ago | MIT | A | 1,252 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,649 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| vscode model-prompts.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 2,022 | GitHub Copilot |