clawrun CLAUDE.md

clawrun CLAUDE.md is an instructions file for coding agents from clawrun-sh/clawrun. It costs 3,617 tokens per session, scanned A, original, Apache-2.0.

Repository instructions for ClawRun, a service that hosts existing open-source AI agents in secure sandboxes and manages their startup, wake-up, snapshots, and shutdown. It does not create the agents itself.

In plain words
What is it for?
Use them when developing deployment commands, sandbox lifecycle behavior, server features, storage, provider integrations, or agent adapters.
Why use it?
They explain the system's lifecycle model and package responsibilities so changes work across the command-line tool, server, runtime, and agent adapters.

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/clawrun-sh/clawrun/claude-md
Clone the repo
git clone --depth 1 https://github.com/clawrun-sh/clawrun

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.

agentmods badge for clawrun CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/clawrun-sh/clawrun/claude-md.svg)](https://agentmods.dev/instructions/clawrun-sh/clawrun/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/clawrun-sh/clawrun/claude-md"><img src="https://agentmods.dev/badge/instructions/clawrun-sh/clawrun/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,617 This file is loaded in full into every session.
When invoked 3,617 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.03617 $0.03617
Opus 5 $0.01809 $0.01809
Sonnet 5 $0.00723 $0.00723
Haiku 4.5 $0.00362 $0.00362

Measured 4d ago against content hash 1f09d526d358, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

clawrun CLAUDE.md scanned grade A with 1 finding 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 4d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

# curl "https://api.telegram.org/bot<TOKEN>/setWebhook?url=<NGROK>/api/v1/webhook/telegram&secret_token=<SECRET>"
CLAUDE.md · 295 lines

How it starts

The opening of the file, as written. The whole thing — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — ClawRun

What This Is

ClawRun is a hosting and lifecycle layer for open-source AI agents. It deploys agents into secure sandboxes (Vercel Sandbox, with more providers coming) and manages their full lifecycle, including startup, heartbeat keep-alive, snapshot/resume, graceful shutdown, and wake-on-message.

ClawRun does NOT build its own agent. It hosts existing ones. ZeroClaw is the default agent; the Agent interface makes it pluggable.

The architecture is a persistent sandbox model: the agent daemon runs continuously inside a sandbox, receives messages via its own HTTP server, and ClawRun manages the sandbox TTL, snapshotting, and wake/sleep cycle around it.

Repo Structure

clawrun/
├── CLAUDE.md
├── AGENTS.md
├── packages/
│   ├── cli/              # `clawrun` CLI — deploy, manage, connect to instances
│   ├── server/           # Next.js app: API routes, handlers, UI (deployed per instance)
│   ├── runtime/          # Sandbox lifecycle, sidecar, storage, config
│   ├── agent/            # Agent interface + registry (abstract)
│   ├── agent-zeroclaw/   # ZeroClaw agent adapter (implements Agent)
│   ├── provider/         # SandboxProvider interface (abstract)
│   ├── provider-vercel/  # Vercel Sandbox provider (implements SandboxProvider)
│   ├── channel/          # WakeHookAdapter interface + 6 channel adapters
│   ├── sdk/              # Programmatic API — instance management, deploy, client
│   ├── auth/             # JWT auth primitives (jose)
│   ├── logger/           # Structured logging (consola)
│   ├── ui/               # React UI component library (Radix, shadcn, Tailwind)
│   ├── zeroclaw/         # ZeroClaw TS wrapper (binary mgmt, config gen, command building)
│   ├── website/          # Documentation site (Next.js, not published)
│   └── tsconfig/         # Shared TS config (private, not published)
├── presets/
│   └── starter/          # Default preset: ZeroClaw + Vercel
└── notes/                # Architecture notes and comparisons

Read the full file on GitHub · 295 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. 4d ago First seen · 295 lines · 3,617 tokens per session scan A 1f09d526d358

Subscribe to this mod's changes

clawrun CLAUDE.md is an instructions file published in the GitHub repository clawrun-sh/clawrun (94 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 3,617 tokens to every session, about $0.0181 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories