lens-sandbox: Instructions file for Claude Code

CLAUDE.md

lens-sandbox CLAUDE.md is an instructions file for Claude Code from lensapp/lens-sandbox. It costs 9,111 tokens per session, scanned C, original, Apache-2.0.

A project guide for Lens Sandbox, a local desktop app that runs AI agents, commands, and OCI container images inside a microVM, a small isolated virtual computer. It explains how access rules, connectors, and secrets are organized.

In plain words
What is it for?
Use it when changing sandbox behavior, approvals, network rules, connectors, credentials, or platform support on macOS and Linux. It directs contributors to the user documentation and the relevant configuration files.
Why use it?
It helps agents preserve the app's security boundaries and use the correct source of truth for product terms and behavior. It separates project rules, connector permissions, and secret values so sensitive data is not committed accidentally.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions Claude Code.

This is lensapp/lens-sandbox's own configuration. It tells Claude Code how to work on lens-sandbox itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything lens-sandbox configures →

Reuse

Borrowing it

Nothing to install: this file belongs to lensapp/lens-sandbox. 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/lensapp/lens-sandbox/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/lensapp/lens-sandbox

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.

agentmods badge for lens-sandbox CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/lensapp/lens-sandbox/claude-md.svg)](https://agentmods.dev/instructions/lensapp/lens-sandbox/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/lensapp/lens-sandbox/claude-md"><img src="https://agentmods.dev/badge/instructions/lensapp/lens-sandbox/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 9,111 This file is loaded in full into every session.
When invoked 9,111 The same file — it is already loaded in full.
Security scan C 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.09111 $0.09111
Opus 5 $0.04555 $0.04555
Sonnet 5 $0.01822 $0.01822
Haiku 4.5 $0.00911 $0.00911

Measured 3d ago against content hash 0d658d40c064, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade C, and why

lens-sandbox CLAUDE.md scanned grade C 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 3d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

make clean rm -rf bin/ + cargo clean (coverage artifacts live inside target/)
CLAUDE.md · 276 lines

How it starts

The opening of the file, as written. The whole thing — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Product Vision

LNS is a local desktop app for running AI agents, commands, OCI images, and other untrusted workloads inside a microVM. Tagline: "The sandbox you'll actually leave running." One-liner: "Run AI agents, commands, and OCI images locally. Control access into and out of the sandbox."

  • Start here: docs/README.md — the first-party user documentation index.
  • Product language: the Product Vision above and docs/ are the source of truth for terminology and framing; keep naming consistent with them. docs/ is user-facing documentation only — no internal sales/marketing material lives in this repo.
  • Concepts: approvals drive policy authoring; a per-run decisions file (~/.lns/runs/<RUN>/decisions.yaml, a kind: mixin document created empty with the run; a destination no rule decides is asked about) holds the network rules, and lns sandbox save writes a run out as a document to keep; Vz on macOS / KVM on Linux is the only runtime; real secrets stay outside the workload.
  • Target format: docs/sandbox-spec.md is the normative specification for the lns.run/v1 document format and the decisions behind it. The code does not implement all of it — see Transitional mode before you touch a document-format surface.
  • Sibling product: Lens Agents is the centrally managed counterpart for IT teams. Same policy model.

Before proposing new features or architecture, consider whether they preserve the core principles: a sandbox you don't turn off, disposable, never leaky (a run's state never escapes the sandbox; stopped runs persist until removed, and one prune sweeps them all away), no system dependencies (the user runs one binary; no apt/brew preflight, no privileged installer), policy you run into, not write, the working directory only roots relative paths, real secrets stay outside the workload. A small user-launched background service (the tray-resident lns-service, started by lns service start and stoppable via the tray Quit menu or lns service stop) is part of "a sandbox you don't turn off" — not a daemon in the apt/launchd sense.

Read the full file on GitHub · 276 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. 3d ago Changed · +40 lines · +1,460 tokens per session 0d658d40c064
  2. 4d ago Changed · -1 lines · -87 tokens per session 71d35d4d3aef
  3. 8d ago First seen · 237 lines · 7,738 tokens per session scan C 252ff57224f2

Subscribe to this mod's changes

lens-sandbox CLAUDE.md is an instructions file published in the GitHub repository lensapp/lens-sandbox (10 stars, last pushed today), licensed Apache-2.0. It adds 9,111 tokens to every session, about $0.0456 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

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

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens