What is agents-at-scale-ark CLAUDE.md?

A project instruction guide for the Ark codebase, covering its folder layout, coding rules, and checks that must pass before code is pushed. It includes guidance for Go, Python, and TypeScript projects.

Use it when editing Ark code, finding the right project folders, or preparing changes for a push.

It gives an agent the repository-specific rules needed to make changes without breaking formatting, linting, tests, or CI checks.

How mckinsey/agents-at-scale-ark configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to mckinsey/agents-at-scale-ark. 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/mckinsey/agents-at-scale-ark/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/mckinsey/agents-at-scale-ark

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/mckinsey/agents-at-scale-ark/claude-md/github.svg)](https://agentmods.dev/instructions/mckinsey/agents-at-scale-ark/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/mckinsey/agents-at-scale-ark/claude-md"><img src="https://agentmods.dev/badge/instructions/mckinsey/agents-at-scale-ark/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/mckinsey/agents-at-scale-ark/claude-md"><img src="https://agentmods.dev/badge/instructions/mckinsey/agents-at-scale-ark/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,628 This file is loaded in full into every session.

When invoked 2,628 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	$0.02628	$0.02628
Opus 5.5	$0.01051	$0.01051
Sonnet 5.5	$0.00526	$0.00526
Haiku 4.5	$0.00263	$0.00263

Measured 2mo ago against content hash 31baad86c1d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agents-at-scale-ark 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 2mo 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 · 271 lines

How it starts

The opening of the file, as written. The whole thing — 271 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

NEVER add comments to generated code unless explicitly requested by the user

Pre-Push Gates (Non-Negotiable)

BEFORE pushing ANY commit, make lint and make test MUST pass locally in every directory the change touches. Exact commands per stack live in "Build Instructions" below; do not skip them.

Never push with lint failures. Same rules CI enforces (golangci-lint + gofumpt for Go, ruff/pyright for Python, eslint for TypeScript). Local pass is the minimum bar.
Never push with failing tests.
Never bypass hooks (--no-verify, --no-gpg-sign) unless the user explicitly asks.
"One-line changes" hide gofumpt and whitespace diffs the most. Run the gates regardless of change size.
Tooling gotcha (Go): GOLANGCI_LINT_VERSION in ark/Makefile may lag local Go. If make lint errors with "Go language version ... used to build golangci-lint is lower than the targeted Go version", fall back to gofumpt -l . (install via go install mvdan.cc/gofumpt@latest) — this catches the formatting rule that breaks CI most often. Full golangci-lint still runs in CI.

Project Structure

Core Folders

ark/ - Kubernetes operator (Go)
- Controller reconciles CRDs: Agent, Model, Query, Team, MCPServer, ExecutionEngine, A2AServer
- Webhooks for validation and mutation (including migration warnings)
- executors/completions/ - Default executor (separate deployment, communicates with controller via A2A)
- The controller dispatches queries to the appropriate executor via A2A protocol
lib/ark-sdk/ - Python SDK (generated + overlay)
- Generated from CRDs via OpenAPI, with hand-written overlay for executor interfaces
- BaseExecutor ABC and ExecutorApp (A2A bridge) provide the standard interface for pluggable executors
- Downstream executor implementations live in the marketplace
services/ - Component services
- ark-api/ - REST API gateway (Python/FastAPI) with streaming, A2A, broker integration
- ark-broker/ - In-memory event bus (Node.js/Express) for messages, chunks, traces, events, sessions
- ark-dashboard/ - Web UI (Next.js/React)
- ark-mcp/ - MCP server host service
- localhost-gateway/ - Local development gateway

Read the full file on GitHub · 271 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.

2mo ago First seen · 271 lines · 2,628 tokens per session scan A 31baad86c1d1

Subscribe to this mod's changes

agents-at-scale-ark CLAUDE.md is an instructions file published in the GitHub repository mckinsey/agents-at-scale-ark (434 stars, last pushed yesterday), licensed Apache-2.0. It adds 2,628 tokens to every session, about $0.0105 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 2026-09-03.

CI and CD Cloud DevOps and cloud Git and workflow agentic-ai ai kubernetes productid-20863

agents-at-scale-ark CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
agents-at-scale-ark CLAUDE.mdmckinsey/agents-at-scale-ark	Instructions file	434	yesterday	Apache-2.0	A	2,628	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

agents-at-scale-ark: Instructions file for Claude Code