agentkit CLAUDE.md

A project guide for Worldcoin AgentKit, a toolkit for building AI agents that use World ID and World Chain. It documents the repository, its commands, and its checks.

In plain words
What is it for?
Working in the AgentKit monorepo, including its TypeScript library, command-line tool, and Solidity smart contracts.
Why use it?
It gives coding agents the project context and exact commands needed to install dependencies, build, test, format, and check the code.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/worldcoin/agentkit/claude-md
Clone the repo
git clone --depth 1 https://github.com/worldcoin/agentkit
Per session 702 This file is loaded in full into every session.
When invoked 702 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00702 $0.00702
Opus 5 $0.00351 $0.00351
Sonnet 5 $0.00140 $0.00140
Haiku 4.5 $0.00070 $0.00070

Measured 2d ago against content hash bb5f6df568b3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentkit 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 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.

CLAUDE.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Project Overview

Monorepo for the Worldcoin AgentKit — a toolkit for building AI agents that integrate with World ID and World Chain.

  • agentkit/ — Core TypeScript library (@worldcoin/agentkit). Handles signature verification, x402 payments, AgentBook lookups, and agent registration validation.
  • cli/ — CLI tool (@worldcoin/agentkit-cli) for agent registration via World ID.
  • contracts/ — Solidity smart contracts (AgentBook registry). Built with Foundry.
  • skills/ — Claude AI integration skills for agentkit.

Dev Commands

# Install dependencies
bun install

# Build the library
bun run build:agentkit

# Run TypeScript tests (from agentkit/)
cd agentkit && bun test

# Run Solidity checks (from contracts/)
cd contracts && forge build && forge test -vvv

# Format code
npx prettier --write .

Pre-PR Checklist

CI only runs Solidity checks — TypeScript issues must be caught locally.

  1. bun run build:agentkit — must compile cleanly
  2. cd agentkit && bun test — all tests must pass
  3. npx prettier --write . — format all files
  4. If touching contracts: cd contracts && forge fmt && forge build && forge test -vvv

Code Style

Prettier handles formatting (config in root package.json), but know the conventions:

  • Tabs for indentation, no semicolons, single quotes
  • 120 character line width
  • Trailing commas in ES5 positions
  • Imports sorted descending via prettier-plugin-sort-imports-desc

Naming

  • camelCase for functions and variables
  • PascalCase for types and interfaces
  • kebab-case for file names
  • SCREAMING_SNAKE_CASE for constants

Exports

  • Named exports only — no default exports
  • Barrel file at agentkit/src/index.ts grouped by category (constants, types, server, chain utilities, etc.)
  • New public exports must be added to the barrel file

TypeScript

  • Strict mode is enabled — do not weaken it
  • Use the type keyword for type-only imports: import type { Foo } from './bar'
  • Zod schemas co-located with their types, derive types via z.infer<typeof Schema>
  • Bun native test runner (bun:test) — not Jest, not Vitest

Read the full file on GitHub · 88 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.

  1. 2d ago First seen · 88 lines · 702 tokens per session scan A bb5f6df568b3

Subscribe to this mod's changes

agentkit CLAUDE.md is an instructions file published in the GitHub repository worldcoin/agentkit (23 stars, last pushed 5d ago), licensed MIT. It adds 702 tokens to every session, about $0.0035 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.