What is railroaded CLAUDE.md?

A project instruction file for Railroaded, an online game where AI players and an AI Dungeon Master run simplified Dungeons & Dragons games without human control during play.

Use it when adding game features, fixing bugs, or consulting the documented mechanics, architecture, agent tools, and coding practices.

It gives the coding agent the game’s rules, architecture, and development conventions for an existing TypeScript project, so changes fit the current system.

How kimosahy/railroaded configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to kimosahy/railroaded. 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/kimosahy/railroaded/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/kimosahy/railroaded

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/kimosahy/railroaded/claude-md/github.svg)](https://agentmods.dev/instructions/kimosahy/railroaded/claude-md)

Your own site

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

Per session 5,023 This file is loaded in full into every session.

When invoked 5,023 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	$0.05023	$0.05023
Opus 5	$0.02511	$0.02511
Sonnet 5	$0.01005	$0.01005
Haiku 4.5	$0.00502	$0.00502

Measured 9d ago against content hash fe515565d8a6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

railroaded 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 9d 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 · 369 lines

How it starts

The opening of the file, as written. The whole thing — 369 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Railroaded — Game Design Specification

Why is this file called CLAUDE.md? It's the instruction file for Claude Code, Anthropic's AI coding agent. Claude Code reads this file at the start of every development session to understand the game's architecture, rules, and constraints before implementing features or fixing bugs. It's the source of truth for the entire codebase.

You are working on Railroaded — an autonomous AI D&D platform where AI agents play D&D with no humans in the loop during gameplay. AI players + AI Dungeon Master, server handles all rules and dice.

This is NOT a greenfield project. v1 is fully built, deployed, and has been playtested. You are adding features and fixing bugs in an existing ~8,700 line TypeScript codebase.

Reference Docs

Read these when you need depth. Don't memorize — look things up.

Game rules: docs/game-mechanics.md — D&D 5e simplified. Races, classes, combat, spells, equipment, resting, leveling.
Architecture: docs/architecture.md — Data models, API design, project structure, agent tool interface, deployment.
Working patterns: docs/cc-patterns.md — Commit discipline, testing, TypeScript rules, how to work with existing code, bug patterns to avoid.
Known issues: docs/known-issues.md — All bugs and gaps from playtesting, prioritized.

Current Sprint: Sprint D — Spectator Experience & Documentation

Goal: Transform the spectator experience from a data dashboard into theater. Agent-first design — agents initiate sessions, no scheduler. All frontend pages overhauled with consistent design language.

Done (Sprint C — Persistence & World)

Monster turn resolution — monster_attack tool works, initiative auto-advances through monsters, DM can resolve monster turns.
Event persistence — logEvent() writes to both in-memory array and session_events DB table. Character snapshots at session-end and combat-end.
Party names — Procedural generator using race/class composition. Stored in DB, surfaced in all party endpoints.
Load from DB on restart — loadPersistedState() rebuilds characters/parties/events from DB at startup.
Narrator layer — narrations table, POST /narrator/narrate (auth'd), GET /spectator/narrations (public).
Homepage heartbeat — "Latest from the Dungeons" narration feed on index.html with auto-refresh, XSS-safe.
WebSocket turn notifications — notifyTurnChange() pushes your_turn to players/DM.
Bonus actions + reactions — TurnResources tracking, bonus_action tool, reaction tool, end_turn tool.
Death saves with drama — WebSocket broadcasts on every death save result, nat 20 revival announcements.

Read the full file on GitHub · 369 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.

9d ago First seen · 369 lines · 5,023 tokens per session scan A fe515565d8a6

Subscribe to this mod's changes

railroaded CLAUDE.md is an instructions file published in the GitHub repository kimosahy/railroaded (2 stars, last pushed 12d ago), licensed MIT. It adds 5,023 tokens to every session, about $0.0251 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.

Autonomous Autonomous Agents Game development ai-agents dnd dnd5e mcp

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

railroaded: Instructions file for Claude Code