agent-world CLAUDE.md

Repository instructions for AgentWorld, a framework for simulating several AI agents in configurable virtual worlds. They explain the project structure, communication style, and development preferences.

In plain words
What is it for?
Use them when developing or documenting AgentWorld. They help with TypeScript project work, plugin-based simulation changes, and communication with the project maintainer.
Why use it?
They give coding agents the project context needed to make changes that fit the repository. They also clarify language, architecture, and compatibility expectations.

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/nicepkg/agent-world/claude-md
Clone the repo
git clone --depth 1 https://github.com/nicepkg/agent-world
Per session 3,291 This file is loaded in full into every session.
When invoked 3,291 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 $0.03291 $0.03291
Opus 5 $0.01646 $0.01646
Sonnet 5 $0.00658 $0.00658
Haiku 4.5 $0.00329 $0.00329

Measured 2d ago against content hash 31ccdc5bfa0c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-world 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 · 262 lines

How it starts

The opening of the file, as written. The whole thing — 262 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentWorld — Multi-Agent Simulation Framework

Everything-as-Plugin architecture for multi-agent simulation. Change the config, run a completely different world.

User Profile

  • Name: Jinming Yang
  • Role: nicepkg open source org founder, framework author
  • Goal: Build a mind-blowing AI multi-agent simulation open source project + Bilibili videos
  • Style: Rapid iteration — get it working first, then optimize
  • Tech Background: TypeScript full-stack, frontend origin

Communication & Language

  • Chinese for communication, English for code, comments, docs, and technical terms
  • Be specific and actionable — no vague advice
  • Show code examples, not just descriptions
  • No backward compatibility needed — the project is NOT open-sourced yet. Delete old code directly, never deprecate.

Project Structure

agent-world/
├── CLAUDE.md                          # This file (system prompt for Claude)
├── package.json                       # pnpm monorepo root (turbo + vitest + husky)
├── pnpm-workspace.yaml                # Workspace config + dependency catalog
├── turbo.json                         # Turbo task pipeline
├── vitest.config.ts                   # Root-level vitest config
├── eslint.config.mjs                  # ESLint flat config (v10)
├── commitlint.config.js               # Conventional commits
├── packages/
│   ├── core/                          # @agent-world/core — Engine, EventBus, Plugin system
│   ├── brain-llm/                     # @agent-world/brain-llm — LLM brain (via Vercel AI SDK)
│   ├── memory/                        # @agent-world/memory — Memory plugins (Simple + Tiered)
│   ├── world/                         # @agent-world/world — World plugins (Grid, Abstract)
│   ├── renderer-server/               # @agent-world/renderer-server — WebSocket renderer (server)
│   ├── renderer-client/               # @agent-world/renderer-client — WebSocket client (browser, zero-dep)
│   └── ui/                            # @agent-world/ui — React hooks, components, primitives, providers, server utils
├── examples/
│   ├── hello-world/                   # Minimal 5-agent grid demo
│   ├── spring-autumn/                 # AI Spring & Autumn Period — 7 kingdoms scenario
│   └── template/                      # Copy-this-to-start experiment scaffold
├── docs/                              # Agent output directories + design docs
└── .claude/
    ├── agents/                        # 10 expert agent personas
    └── skills/                        # Skill definitions

Read the full file on GitHub · 262 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 · 262 lines · 3,291 tokens per session scan A 31ccdc5bfa0c

Subscribe to this mod's changes

agent-world CLAUDE.md is an instructions file published in the GitHub repository nicepkg/agent-world (5 stars, last pushed 6mo ago), licensed MIT. It adds 3,291 tokens to every session, about $0.0165 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 2026-08-31.