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.
npx agentmods add instructions/risingwavelabs/box0/claude-mdgit clone --depth 1 https://github.com/risingwavelabs/box0Wrote 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.
[](https://agentmods.dev/instructions/risingwavelabs/box0/claude-md)<a href="https://agentmods.dev/instructions/risingwavelabs/box0/claude-md"><img src="https://agentmods.dev/badge/instructions/risingwavelabs/box0/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01642 | $0.01642 |
| Opus 5 | $0.00821 | $0.00821 |
| Sonnet 5 | $0.00328 | $0.00328 |
| Haiku 4.5 | $0.00164 | $0.00164 |
Grade A, and why
box0 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Project
Box0 is a multi-agent platform. It lets you run multiple AI agents in parallel across machines. Rust codebase, single binary (b0). npm package: @box0/cli.
Conventions
- Never use em dashes. Use periods, commas, colons, or "- " instead.
- All content in English.
- README code blocks must be copy-paste safe. No inline comments on the same line as commands. No blocking commands (like
b0 server) in the same block as other commands. Use blockquotes for conversation examples, not code blocks. - Always test changes before committing. Run
cargo testat minimum (unit + integration tests). For user-facing features, runtests/e2e.shwhich requires Claude Code or Codex. - After code changes, always update README.md, CLAUDE.md, docs/, and the skill content in config.rs if affected. This is critical.
- SKILL.md (repo root) is the canonical skill definition. Any change to CLI commands, flags, or agent behavior must be reflected in it.
- Commit messages: imperative mood, concise first line, details in body.
- No documentation files unless explicitly requested.
Architecture
src/lib.rs- Library crate, re-exports all modulessrc/main.rs- CLI entry point, all subcommand dispatchsrc/server.rs- Axum HTTP server, route handlers, auth middleware,build_router()for testssrc/db.rs- SQLite schema, models, all queriessrc/daemon.rs- Daemon (local + remote), processes agent inboxes, spawns runtimesrc/client.rs- HTTP client for CLI-to-server communicationsrc/config.rs- Server config, CLI config, skill installation, pending statesrc/scheduler.rs- Cron job scheduler, runs recurring tasks on interval
Resource model
- Workspaces are logical groups for organizing agents, tasks, and team access.
- Agents belong to a workspace. Workspace controls visibility.
Auth model
- Users have unique keys. Keys identify users, not workspaces.
- Each user gets a personal workspace on creation.
- Users can be in multiple workspaces.
--workspaceflag selects which workspace to operate in. Defaults todefault_workspacein config. - Agents track
registered_by. Only the creator can remove/update/stop their agents. - Admin user is created on first server start. Server auto-writes CLI config and sets
default_workspaceautomatically. No login command needed.
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.
- 6d ago First seen · 102 lines · 1,642 tokens per session scan A 3593e4af1566
box0 CLAUDE.md is an instructions file published in the GitHub repository risingwavelabs/box0 (81 stars, last pushed 5mo ago), licensed MIT. It adds 1,642 tokens to every session, about $0.0082 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.
Other instructions, from other repositories
commonly CLAUDE.md
Claude Code instructions for Team-Commonly/commonly, covering claude.md / agents.md, 🧠 product vision & architecture philosophy, what commonly is, the architecture model and key concepts.
anomalia AGENTS.md
AGENTS.md instructions for anomaliaso/anomalia, covering anomalia — social media ai autopilot, cli, install (standalone binary, no runtime needed), quick commands and architecture (this repo — the server side of the cli).
commonly AGENTS.md
AGENTS.md instructions for Team-Commonly/commonly, a project described as: Open-source room for humans + cross-vendor AI agents. Every agent gets its own name, memory, skills, and workstation. Any runtime, your infra — no per-agent fees.
anomalia CLAUDE.md
Claude Code instructions for anomaliaso/anomalia, covering anomalia — social media ai autopilot, cli, install (standalone binary, no runtime needed), quick commands and architecture (this repo — the server side of the cli).
Assistant CLAUDE.md
Claude Code instructions for bearlike/Assistant, covering agents guide — mewbo, mandatory: hydrate before touching files, what mewbo is, monorepo layering — read before adding a module and engineering principles.
PuPu CLAUDE.md
Claude Code instructions for haoxiang-xu/PuPu, covering gitnexus — code intelligence, always do, never do, resources and cli.