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/miantiao-me/cloud-claw/agents-mdgit clone --depth 1 https://github.com/miantiao-me/cloud-clawWrote 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/miantiao-me/cloud-claw/agents-md)<a href="https://agentmods.dev/instructions/miantiao-me/cloud-claw/agents-md"><img src="https://agentmods.dev/badge/instructions/miantiao-me/cloud-claw/agents-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 | $0.01532 | $0.01532 |
| Opus 5 | $0.00766 | $0.00766 |
| Sonnet 5 | $0.00306 | $0.00306 |
| Haiku 4.5 | $0.00153 | $0.00153 |
Grade A, and why
cloud-claw AGENTS.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 5d 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 — 170 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Development Guidelines
Development guidelines for AI coding agents working on Cloud Claw.
Project Overview
Cloud Claw is a TypeScript project running on Cloudflare Workers + Containers. It uses Durable Objects with @cloudflare/containers to manage containerized AI assistant workloads.A Worker handles routing/auth, forwards requests to a singleton container running an OpenClaw gateway instance, and proxies Chrome DevTools Protocol (CDP) sessions via Cloudflare Browser bindings.
Tech Stack: Cloudflare Workers, TypeScript (ES2024), pnpm (v10.28.2)
Commands
pnpm install # Install dependencies
pnpm dev # Start local dev server (binds 0.0.0.0)
pnpm deploy # Deploy to Cloudflare
pnpm lint # Run formatter (oxfmt) + linter (oxlint)
pnpm cf-typegen # Regenerate Cloudflare type definitions
npx oxfmt # Format only
npx oxlint # Lint only
No test suite configured. Validate changes with pnpm lint and pnpm dev.
Project Structure
src/
├── index.ts # Workers entry point (ExportedHandler), routing, basic auth
├── container.ts # AgentContainer class (extends Container), WebSocket gateway
└── cdp.ts # Chrome DevTools Protocol proxy (chunked binary WebSocket framing)
wrangler.jsonc # Wrangler config (containers, bindings, placement)
tsconfig.json # TypeScript config (ES2024, strict, bundler resolution)
Dockerfile # Container image: OpenClaw gateway + TigrisFS S3 mount
worker-configuration.d.ts # Auto-generated Cloudflare bindings (DO NOT EDIT)
.oxfmtrc.json # Formatter config (single quotes, no semicolons, spaces)
Code Style
Formatting (oxfmt — .oxfmtrc.json)
- Single quotes, no semicolons, spaces for indentation (oxfmt overrides
.editorconfig) - LF line endings, always insert final newline
- Run
pnpm lintbefore committing
TypeScript
- Target: ES2024, Module: ES2022, Resolution: Bundler
- Strict mode: Enabled — No emit (Wrangler bundles)
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.
- 5d ago First seen · 170 lines · 1,532 tokens per session scan A 880ef63bed3d
cloud-claw AGENTS.md is an instructions file published in the GitHub repository miantiao-me/cloud-claw (262 stars, last pushed 6mo ago), licensed MIT. It adds 1,532 tokens to every session, about $0.0077 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
openclaw-memory-vectorize AGENTS.md
Instructions for Atlas-Os1/openclaw-memory-vectorize, covering agents.md, quick setup (execute these commands), 1. install dependencies, 2. create vectorize index (skip if exists) and 3. deploy worker.
openclaw-kubernetes AGENTS.md
Instructions for feiskyer/openclaw-kubernetes, covering agents.md, links, development commands, lint the chart against all values files and render templates to verify correctness (output discarded).
autoclaw.sh AGENTS.md
Instructions for datopian/autoclaw.sh, covering agent instructions, quick reference, landing the plane (session completion), issue tracking with bd (beads) and why bd?.
openclaw-kubernetes CLAUDE.md
Instructions for feiskyer/openclaw-kubernetes, covering claude.md, links, development commands, lint the chart against all values files and render templates to verify correctness (output discarded).
OpenClaw-ios CLAUDE.md
Claude Code instructions for Parham-dev/OpenClaw-ios, covering claude.md, build & run, architecture, layer flow and key abstractions.
librefang AGENTS.md
AGENTS.md instructions for librefang/librefang, covering stack, layout, build, architecture and kernelhandle trait.