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 skills add WaHaiLong/KingdeeMCP --skill kingdee-mcp-devgit clone --depth 1 https://github.com/WaHaiLong/KingdeeMCPWrote 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/wahailong/kingdeemcp/kingdee-mcp-dev)<a href="https://agentmods.dev/skills/wahailong/kingdeemcp/kingdee-mcp-dev"><img src="https://agentmods.dev/badge/skills/wahailong/kingdeemcp/kingdee-mcp-dev/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/wahailong/kingdeemcp/kingdee-mcp-dev"><img src="https://agentmods.dev/badge/skills/wahailong/kingdeemcp/kingdee-mcp-dev.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 | $0.00074 | $0.01529 |
| Opus 5 | $0.00037 | $0.00764 |
| Sonnet 5 | $0.00015 | $0.00306 |
| Haiku 4.5 | $0.00007 | $0.00153 |
Grade A, and why
kingdee-mcp-dev 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 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.
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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
金蝶 MCP 开发团 · 知识库
本 Skill 是 kingdee-mcp-dev 专家团的共享知识底座，供主理人（龚联达）与四位团员（范探源 / 寇豆码 / 严过关 / 温书成）在协作时查阅。目标是把金蝶开放平台 ApiDoc 的全部接口集成进 D:\AI\projects\kingdee-mcp。
项目事实（已实证，勿凭记忆）
- 源码位置：
D:\AI\projects\kingdee-mcp，Python + FastMCP。 - 工具规模：
src/kingdee_mcp/server.py当前 97 个@mcp.tool（2026-08-07 实测 97；本日新增 6 个标准动作通用工具kingdee_cancel_bills/kingdee_void_bills/kingdee_close_bill/kingdee_unclose_bill/kingdee_forbid_bills/kingdee_enable_bills，覆盖 14/15 标准动作；报表由kingdee_query_report覆盖；覆盖 13 大业务域 + 4 个 SQL Server 探查）。 - 构建后端：
hatchling（python -m build），dist/产物kingdee_mcp-0.2.0-py3-none-any.whl/.tar.gz。 - 运行环境：金蝶 K3Cloud
http://<K3Cloud-Server>/k3cloud/，账套<ACCT_ID>（<公司名>）。 - 登录方式：仅账号密码（
ValidateUser），无第三方应用授权；配置KINGDEE_SERVER_URL/KINGDEE_ACCT_ID/KINGDEE_USERNAME/KINGDEE_PASSWORD。 - WebAPI 权限拒绝返回
[[{'Result':{报错}}]]（r[0][0] 是 dict），正常数据返回[[值,...]]（r[0][0] 是标量）—— 写查询脚本务必用此区分，否则会把错误当数据。
ApiDoc 结构（浏览器实测）
https://openapi.open.kingdee.com/ApiDoc 是 Element UI(el-tree) SPA，需真实浏览器爬取。结构为 16 业务领域 → 77 模块 → 操作（标准金蝶 WebAPI 动作集）：
- 16 业务领域：员工服务、财务会计、税务管理、成本管理、资产管理、管理会计、供应链、电商与分销、零售管理、生产制造、质量管理、星空云服务、基础管理、BOS、移动应用、PLM。
- 标准操作动作（每个单据模块通用）：保存 / 提交 / 审核 / 反审核 / 撤销 / 下推 / 作废 / 整单关闭 / 反关闭 / 批量保存 / 单据查询 / 查看 / 暂存 / 删除，外加部分领域专属动作（如费用类「申请单退款」）。
- 原始抓取树见
references/api-inventory-raw.json（领域→模块已全量；操作级仅在「员工服务」域完整抓取，其余域操作需逐模块切换抓取）。
集成工作流（团队 SOP 映射）
| 阶段 | 负责团员 | 关键动作 | 输入 → 输出 |
|---|---|---|---|
| 探字段 / WebAPI 映射 | 范探源 | kingdee_get_fields 探 FormId 与字段 Key；映射 ApiDoc 操作到 Save/Submit/Audit/Push/ExecuteBillQuery |
任务 → 字段/操作映射表 |
实现 @mcp.tool |
寇豆码 | 复用提速工具，落地工具代码 | 映射表 → 工具签名 |
| 测试回归 | 严过关 | bin/kmcp test + 用例；区分权限拒绝 vs 正常数据 |
工具签名 → 回归报告 |
| 文档 / 发布 | 温书成 | 更新 docs/、mcp_optimization_notes.md、CHANGELOG；bin/kmcp build → twine upload → Pages |
变更 → 发布清单 |
提速工具（已落地 server.py，优先复用）
kingdee_validate_bill：保存前校验，返回auto_fixes / missing_required / entry_issues / suggestions，不真正保存。kingdee_get_bill_template：取已验证 model 骨架，支持SAL_Quotation/SAL_SaleOrder/PUR_PurchaseOrder。kingdee_refresh_metadata：强制刷新元数据。- 元数据已落盘缓存到
~/.workbuddy/kingdee_metadata_cache/。
What ships with it
6 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.
- 9d ago First seen · 64 lines · 74 tokens per session scan A f6dcd99b45c3
kingdee-mcp-dev is a skill published in the GitHub repository WaHaiLong/KingdeeMCP (76 stars, last pushed 28d ago), licensed MIT. It adds 74 tokens to every session and 1,529 once invoked, about $0.0004 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 skills, from other repositories
MCP Server Builder
Step-by-step guidance for creating a new MCP server with FastMCP or the TypeScript SDK — tool definitions, resource handlers, error responses, and usage examples.
smplkit
Use smplkit when writing application code that needs feature flags, runtime configuration, dynamic log levels, or audit events — and to operate that same platform from your agent. Two halves that compose. (1) AUTHOR with the smplkit SDKs (Python, TypeScript, Go, Java, C#, Ruby): resolve a feature flag, read a config…
kingdee-agent-rules
A set of usage rules for AI agents that connect to Kingdee Cloud Starry Sky, also known as K3Cloud, an enterprise business system.
event-store-design
Design and implement event stores for event-sourced systems. Use when building event sourcing infrastructure, choosing event store technologies, or implementing event persistence patterns.
cqrs-implementation
Implement Command Query Responsibility Segregation for scalable architectures. Use when separating read and write models, optimizing query performance, or building event-sourced systems.
workflow-orchestration-patterns
Design durable workflows with Temporal for distributed systems. Covers workflow vs activity separation, saga patterns, state management, and determinism constraints. Use when building long-running processes, distributed transactions, or microservice orchestration.