What is willys-mcp CLAUDE.md?

A set of Claude Code instructions for the willys-mcp repository, whose code and user-facing text are written in Swedish. It documents commands, conventions, error handling, and known pitfalls.

Use it when editing willys-mcp, running its type check or smoke test, following Swedish-language conventions, and handling failures through the repository's result type.

It gives an agent the repository-specific rules needed to make safe changes and run the correct checks. It also distinguishes local type checks from smoke tests that log in to the real Willys service.

How oscarthroedsson/willys-mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to oscarthroedsson/willys-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/oscarthroedsson/willys-mcp/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/oscarthroedsson/willys-mcp

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/oscarthroedsson/willys-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/oscarthroedsson/willys-mcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/oscarthroedsson/willys-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/oscarthroedsson/willys-mcp/claude-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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/oscarthroedsson/willys-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/oscarthroedsson/willys-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,111 This file is loaded in full into every session.

When invoked 2,111 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.02111	$0.02111
Opus 5.5	$0.00844	$0.00844
Sonnet 5.5	$0.00422	$0.00422
Haiku 4.5	$0.00211	$0.00211

Measured 23d ago against content hash bc56934009a7, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

willys-mcp CLAUDE.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 23d 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.

CLAUDE.md · 133 lines

How it starts

The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Vägledning för Claude Code när du arbetar i det här repot.

Kommandon

npm run typecheck      # tsc --noEmit — enda verkliga porten före commit
npm run mcp            # kör servern på stdio
npm run smoke          # tests/smoke.ts, spawnar servern och kör riktiga anrop
npm run test:volume    # ren matte, inget nätverk
npm run test:coverage  # täckningsoptimeraren, inget nätverk
npm run test:planner   # behovsregistret, inget nätverk (egen plan-id)

Det finns ingen unit-testsvit och inget byggsteg — tsx kör TypeScript direkt. typecheck är därför den enda automatiska kontrollen. Kör den alltid efter en ändring.

npm run smoke kräver giltiga uppgifter i .env och gör riktiga anrop mot Willys. Den loggar in, lägger en vara i kundvagnen och tar bort den igen, och kontrollerar att vagnen är som före.

Språk

Kod, kommentarer och användarvänd text är på svenska. Verktygsbeskrivningar, felmeddelanden och instructions-strängen läses av en modell och ska vara lika precisa som kommentarerna.

Konventioner

Inga kastade undantag i uppströmslagret. Allt som kan misslyckas returnerar Result<T> från core/result.ts. Felklassen (AUTH_EXPIRED, SHAPE, NOT_FOUND, UPSTREAM, INPUT) ska överleva hela vägen till verktygssvaret — den är det modellen agerar på.

En adapter per plattform, en fil per uppströmsresurs. src/upstream/<plattform>/ (axfood/, coop/ …) har en modul per resurs och en barrel i index.ts. Lägg inte tillbaka blandade ansvar i en gemensam klientfil. src/chains/ binder ihop dem: types.ts är ChainAdapter-kontraktet, registry.ts mappar ChainId → adapter, och delad logik som ska gälla alla kedjor (rangordning i cheapest.ts, butiksregistret i stores.ts) ligger där — aldrig i en enskild adapter. Willys och Hemköp är samma Axfood-API; en skillnad mellan dem hör hemma i AxfoodSite, inte i en kopia av koden. API-kartan per kedja står i docs/kedjor.md.

Ett verktyg = ett objekt = en fil. Varje verktyg definieras med defineTool i en gruppfil under src/tools/ (session.ts, search-tools.ts, planner-tools.ts …) och samlar namn, schema, beskrivning och handler — inklusive sin egen svarsformning. Lägg till ett verktyg genom att lägga det i rätt gruppfil; splittra det INTE över namnkonstant + schemaregister + dispatch igen. Namnet deklareras fortfarande i constants/tools.ts, och en startup-vakt i tools/registry.ts kastar om ett deklarerat namn saknar handler.

Read the full file on GitHub · 133 lines

Changes

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.

23d ago Changed · +29 lines · +583 tokens per session bc56934009a7
2mo ago First seen · 104 lines · 1,528 tokens per session scan A 556f250e74c2

Subscribe to this mod's changes

willys-mcp CLAUDE.md is an instructions file published in the GitHub repository oscarthroedsson/willys-mcp (2 stars, last pushed 24d ago), licensed MIT. It adds 2,111 tokens to every session, about $0.0084 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.

willys-mcp CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
willys-mcp CLAUDE.mdoscarthroedsson/willys-mcp	Instructions file	2	24d ago	MIT	A	2,111	Claude Code
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	5,516	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
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	275k	3d ago	MIT	A	4,313	Codex, OpenCode

willys-mcp: Instructions file for Claude Code