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/hithink-tech/financial-api/agents-mdgit clone --depth 1 https://github.com/HiThink-Tech/Financial-APIWhat 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 | $0.01319 | $0.01319 |
| Opus 5 | $0.00660 | $0.00660 |
| Sonnet 5 | $0.00264 | $0.00264 |
| Haiku 4.5 | $0.00132 | $0.00132 |
Grade A, and why
Financial-API 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 2d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
任何 AI coding/data Agent 进入本仓库后的公开入口。项目品牌为“同花顺金融数据服务（hithink finance）”。
必读顺序
- 旧 checkout、旧 Prompt 或旧根级 Python 路径：先读
docs/monorepo-migration.md。 - 所有金融数据任务：先读
skills/hithink-finance/SKILL.md。 - 只加载 Skill 选择的一个接入方式或端点组，不要递归读取全部契约。
Monorepo 边界
hithink-finance-cli/：Node.js CLI 子项目，面向人类、Agent 和自动化；运行时不依赖 Python。python/：唯一 Python 项目根，包含远端取数 toolkit、本地marketdb、示例和测试。docs/：公共文档中心；docs/api/是仓库内唯一上游 REST API 契约源。skills/hithink-finance/：可独立发布的统一 Skill；API/MCP 契约由脚本从docs/镜像。examples/：monorepo 级示例导航与静态灵感。
选择接入方式
| 场景 | 优先入口 |
|---|---|
| 人类终端、Agent 执行、自动化、远端+本地一体化 | hithink-finance CLI |
| 已连接 MCP 的 Chat 场景 | 托管 MCP |
| 零依赖 HTTP、自定义语言或服务端 | REST API |
| Python/Notebook/研究或已有 marketdb | Python toolkit/SDK |
CLI 已安装时先运行 hithink-finance capabilities --format json，再按需运行 schema <id>；MCP 以实时 tools/list 为准；REST 响应字段以 docs/api/ 及远端 https://fuyao.aicubes.cn/llms-full.txt 为准；Python 适配层参数以当前函数签名和 --help 为准。
API Key
所有远端接入方式使用在 https://fuyao.aicubes.cn/admin 获取的统一 API Key。推荐统一来源是用户级 HITHINK_FINANCE_API_KEY，其次是 hithink-finance/credentials.env 用户级凭据文件。
- 每次使用 Skill 时先检查统一凭据来源；找到后直接复用，不得因切换接入方式再次索要。
- 不得强制用户把 Key 粘贴到对话；用户主动提供或选择 Agent 代配时，不得复述，并应安全写入用户级统一凭据来源。
- 不得把 Key 写入代码、Prompt、日志、公开配置、产物或 Git。
- CLI 安装、统一凭据新增或更新后，通过 stdin 登录；已有 CLI 凭据用
auth login --api-key-stdin --replace原子替换。CLI 系统凭据库保留独立副本。 - REST/Python 读取统一环境变量或用户级凭据文件；旧变量仅兼容。
- MCP 优先使用
${HITHINK_FINANCE_API_KEY}，不继承环境时由 Agent 从统一来源配置客户端 Secret。
大数据纪律
禁止把全市场、分页全集、多年或多标的原始结果输出到会话上下文。
<command> ... > /tmp/result.json
# 只报告文件路径、行数、时间窗口和摘要
CLI 优先使用具体命令的 --output、db export 或 market panel --output。Python/marketdb 将结果写到 /tmp/、out/ 或用户指定路径。不要回显凭证或完整数据文件。
文档治理
-
根 README 做项目总览；子目录 README 详细解释当前目录，不把细节继续拆散到不必要的多层文档。
-
上游 REST API 契约只在
docs/api/维护；修改后运行：python scripts/sync_skill_contracts.py python scripts/sync_skill_contracts.py --check
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.
- 2d ago First seen · 89 lines · 1,319 tokens per session scan A 608dcf7fe7b1
Financial-API AGENTS.md is an instructions file published in the GitHub repository HiThink-Tech/Financial-API (2,118 stars, last pushed 5d ago), licensed MIT. It adds 1,319 tokens to every session, about $0.0066 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
china-financial-services AGENTS.md
Instructions for cyijun/china-financial-services, covering agents.md, 项目边界, 目录、marketplace 与唯一数据源, 数据、安全与研究证据 and 修改原则.
aurumq-rl CLAUDE.md
Claude Code instructions for yupoet/aurumq-rl, covering claude.md — aurumq-rl ai 协作约定, 项目定位, 输入数据契约（核心约定）, 技术红线（不可违反） and 1. 训练资源限制.
all-weather-portfolio CLAUDE.md
Claude Code instructions for IdealAuror/all-weather-portfolio, covering claude.md, 策略优化闭环, 前置原则, 第 0 层：哲学基座 and sperandeo 三层结构.
aurumq-rl AGENTS.md
AGENTS.md instructions for yupoet/aurumq-rl, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
china-financial-services CLAUDE.md
Instructions for cyijun/china-financial-services, a project described as: AI agent plugins & managed-agent templates for China A-share market research and financial modeling, powered by Tushare. Research workflows and valuation tools for professional analyst review—not investment advice. | 面向中国 A 股市场的 AI…
stock-sdk AGENTS.md
Instructions for chengzuopeng/stock-sdk, covering stock sdk 项目规范, 规则验证指令, 项目概述, 版本与分支 and 技术栈.