What is agent-dice CLAUDE.md?

A project instruction file for agent-dice, a system that uses probabilistic dice rolls to trigger actions in Claude Code hooks. It documents the project structure, development commands, adapters, installation, and tests.

Use it when developing, installing, or testing agent-dice across Claude Code, Codex, and other supported hosts.

It gives coding agents the project-specific rules and commands needed to change and test agent-dice correctly.

How pro-vi/agent-dice configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to pro-vi/agent-dice. 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/pro-vi/agent-dice/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/pro-vi/agent-dice

Made for: Claude Code, Codex.

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/pro-vi/agent-dice/claude-md/github.svg)](https://agentmods.dev/instructions/pro-vi/agent-dice/claude-md)

Your own site

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

Per session 800 This file is loaded in full into every session.

When invoked 800 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	$0.00800	$0.00800
Opus 5.5	$0.00320	$0.00320
Sonnet 5.5	$0.00160	$0.00160
Haiku 4.5	$0.00080	$0.00080

Measured 2mo ago against content hash b82b507f3577, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

agent-dice 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 · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agent-dice

Probabilistic dice trigger system for Claude Code hooks.

Project Layout

src/index.ts — Public API facade (delegates to core via the Claude adapter)
src/core/ — Host-agnostic engine + DiceHost contract + pure accumulator (no Claude/Bun/fs)
src/adapters/ — Claude Code adapter + trigger renderer; pi/ (in-process host); codex/ (hook-based host — Claude twin: rollout depth parser + host reusing the Claude stores @ ~/.codex/dice)
src/{registry,state,cooldown,transcript,session,roll}.ts — Claude/file primitives
bin/agent-dice.ts — CLI entrypoint
hooks/ — Hook scripts: Claude (stop, session-start) + Codex (codex-stop, codex-session-start)
install.sh — Installer (symlinks, hook registration, dependency checks)
tests/unit/test_cc_dice.bats — BATS test suite
tests/conformance/ — Conformance probes (run.ts entry; facade/storage/boundary/core-engine)
docs/architecture.md — Architecture, design decisions, origin story

Development

bun install                  # install dev dependencies
bun run test                 # run test suite (BATS + conformance). NOT bare `bun test`
./install.sh check           # verify installation
DEBUG=1 bun hooks/stop.ts    # run stop hook with verbose logging

Key Concepts

Core + adapters: src/core/ owns host-agnostic policy behind the DiceHost contract; three hosts ship on it with zero core changes — claude-code.ts (default), codex/ (hook-based, the Claude twin), and pi/ (in-process extension). src/index.ts stays the public Claude facade.
Boundary: nothing under src/core/** may import a Claude/host module, node builtin, Bun, or process.env (enforced by conformance check C8).
Slots: Named dice configurations persisted to ~/.claude/dice/slots.json
Shared pools: checkAllSlots() groups slots by die size, rolls one base die per group
Session isolation: State is per-slot AND per-session via AGENT_DICE_SESSION_ID (or the CC_DICE_SESSION_ID back-compat alias)
Hook exit codes: 0 = silent, 2 = stderr shown to Claude

Read the full file on GitHub · 54 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 · 54 lines · 800 tokens per session scan A b82b507f3577

Subscribe to this mod's changes

agent-dice CLAUDE.md is an instructions file published in the GitHub repository pro-vi/agent-dice (22 stars, last pushed 3mo ago), licensed MIT. It adds 800 tokens to every session, about $0.0032 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 2026-08-30.

agent-dice CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
agent-dice CLAUDE.mdpro-vi/agent-dice	Instructions file	22	3mo ago	MIT	A	800	Claude Code, Codex
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	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	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	2,022	GitHub Copilot

agent-dice: Instructions file for Claude Code