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/marianfoo/sap-mcp-servers/agents-mdgit clone --depth 1 https://github.com/marianfoo/sap-mcp-serversWrote 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/marianfoo/sap-mcp-servers/agents-md)<a href="https://agentmods.dev/instructions/marianfoo/sap-mcp-servers/agents-md"><img src="https://agentmods.dev/badge/instructions/marianfoo/sap-mcp-servers/agents-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 | $0.02497 | $0.02497 |
| Opus 5 | $0.01248 | $0.01248 |
| Sonnet 5 | $0.00499 | $0.00499 |
| Haiku 4.5 | $0.00250 | $0.00250 |
Grade A, and why
sap-mcp-servers 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 6d 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 — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — sap-mcp-servers monorepo
Guidance for AI agents and humans working in this repo. Read this first.
What this repo is
An npm-workspaces monorepo holding three SAP MCP servers plus the shared SAP login module they all build on. Each package is published to npm independently under its own name — the monorepo is for development only.
sap-mcp-servers/
├── package.json # workspaces + root scripts (build order lives here)
├── package-lock.json # single lockfile for the whole repo — do not add per-package lockfiles
├── AGENTS.md # this file
├── README.md # public overview
└── packages/
├── auth/ @marianfoo/sap-mcp-auth — shared SAP IAS/SSO login (Playwright). NO server.
├── api-hub/ sap-api-hub-mcp — MCP server: SAP Business Accelerator Hub
├── roadmap/ sap-roadmap-mcp — MCP server: SAP Road Map Explorer
└── notes/ sap-note-search-mcp — MCP server: SAP Notes / KBAs
Not in this repo: sap-docs (no auth) and arc-1 (live-system tooling) — they live in their own
repos.
The one thing to know: build order
auth must be built before the three servers — they import its compiled types/JS from
@marianfoo/sap-mcp-auth. The root npm run build already does this:
npm run build == build auth → then build api-hub + roadmap + notes
npm install symlinks @marianfoo/sap-mcp-auth into each server from packages/auth (workspace
link). So you can edit the auth module and rebuild without publishing it — the servers pick up
local changes immediately after npm run build (or npm run build:auth). Never npm publish just
to test an auth change locally.
First-time setup
git clone https://github.com/marianfoo/sap-mcp-servers.git
cd sap-mcp-servers
npm install # all workspaces + the auth symlink + one root lockfile
npm run build # auth first, then the three servers
npm run install:browsers # one-time: Chromium for Playwright (also auto-runs via roadmap postinstall)
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.
- 6d ago First seen · 189 lines · 2,497 tokens per session scan A f01fb86c53a6
sap-mcp-servers AGENTS.md is an instructions file published in the GitHub repository marianfoo/sap-mcp-servers (15 stars, last pushed 12d ago), licensed Apache-2.0. It adds 2,497 tokens to every session, about $0.0125 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
btp-cf-mcp AGENTS.md
Instructions for marianfoo/btp-cf-mcp, covering agents.md, project overview, design principles, build & test and live mcp smoke: mcpurl=… mcpkey=… node test/smoke.mjs.
linkedin-mcp-server AGENTS.md
AGENTS.md instructions for prakharagarwal-dev/linkedin-mcp-server, covering repository instructions, product boundary, access and secrets, technical conventions and working style.
freetaxusa-mcp CLAUDE.md
Claude Code instructions for schwarztim/freetaxusa-mcp, covering freetaxusa mcp server, architecture, commands, tool categories and phase 1 (implemented).
sap-adt-mcp CLAUDE.md
Claude Code instructions for engoetz/sap-adt-mcp, covering claude.md, what this is, commands, architecture and invariants that are easy to break.
mcp-recipe-shopping-list AGENTS.md
AGENTS.md instructions for KrivchenkoEgor/mcp-recipe-shopping-list, covering ⛔ критические запреты (читать первым), 1. 🚫 картинки в чате запрещены, 2. 🚫 оба сайта — spa (javascript-heavy), 3. 🚫 не формируем корзину — только список покупок and 4. 🚫 уважение к серверам.
tradingview-mcp AGENTS.md
AGENTS.md instructions for moeuu/tradingview-mcp, covering repository instructions and codex review waiting.