klimkit-remote-control

klimkit-remote-control is a skill for Claude Code, Codex from klimentij/klimkit. It costs 81 tokens per session (2,211 once invoked), scanned B, original, MIT.

A setup and troubleshooting guide for running Claude Code Remote Control on a virtual machine or server. It lets a phone or browser control a coding session that keeps running on that machine.

In plain words
What is it for?
Use it to install, supervise, secure, resize, or debug remote coding sessions and diagnose why they stop after disconnects or reboots.
Why use it?
It helps keep sessions available after a laptop closes and separates memory failures from network disconnects, so the right problem is fixed.

Skill for Claude CodeCodex

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 skills/klimentij/klimkit/klimkit-remote-control
Any agent
npx skills add klimentij/klimkit --skill klimkit-remote-control
Clone the repo
git clone --depth 1 https://github.com/klimentij/klimkit

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.

agentmods badge for klimkit-remote-control

README.md
[![agentmods](https://agentmods.dev/badge/skills/klimentij/klimkit/klimkit-remote-control.svg)](https://agentmods.dev/skills/klimentij/klimkit/klimkit-remote-control)
Your own site
<a href="https://agentmods.dev/skills/klimentij/klimkit/klimkit-remote-control"><img src="https://agentmods.dev/badge/skills/klimentij/klimkit/klimkit-remote-control.svg" alt="Measured on agentmods" height="20"></a>
Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,211 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 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 5.1 $0.00081 $0.02211
Opus 5 $0.00041 $0.01105
Sonnet 5 $0.00016 $0.00442
Haiku 4.5 $0.00008 $0.00221

Measured 5d ago against content hash 519a56bc42c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade B, and why

klimkit-remote-control scanned grade B 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 5d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/healthcheck.sh, scripts/install_remote_control.sh, scripts/preflight.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

sudo scripts/install_remote_control.sh --dir /home/ubuntu/myrepo --name vps-myrepo
skills/klimkit-remote-control/SKILL.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Remote Control on a VM

Make one folder on a machine permanently drivable from a phone or browser. The work executes on that machine — its filesystem, its credentials, its MCP servers — while claude.ai/code is only a window onto it.

The single most common failure is not what people expect. Sessions usually die because the machine ran out of memory and the kernel killed the supervisor, not because the network dropped. Diagnose before you design: a session dying at a memory ceiling and a session dying at a network timeout need different fixes, and only one of them is fixed by Restart=always. See references/oom-and-capacity.md.

Work through the phases in order. Do not install anything before Phase 1 passes — an ineligible account produces a service that restart-loops forever and looks like a bug in your unit file.

Phase 0 — Is this the right tool?

Remote Control drives a session running on this machine. Reach for something else when:

Want Use instead
Work on a repo without local setup, or many parallel cloud tasks Claude Code on the web
Fire work from a chat app (Telegram, Discord, iMessage) on push events Channels
Message a task from the phone and let it spawn a session Dispatch (Desktop)
Run something on a schedule with no session open Scheduled tasks, or cloud routines

Choose Remote Control when execution must stay on this machine — local secrets, a VPC, a database, a GPU, an interactively authenticated CLI — or the job outlives a cloud session.

Phase 1 — Preflight, before installing anything

Run scripts/preflight.sh. It exits non-zero with a specific reason and prints a summary of the machine. Every check exists because it silently breaks the service otherwise; the full rationale is in references/preflight.md.

scripts/preflight.sh --dir /home/ubuntu/myrepo

It verifies, in order: claude is installed and supports remote-control; the login is a claude.ai OAuth session, not an API key and not a setup-token; ANTHROPIC_BASE_URL is unset or api.anthropic.com; the target directory exists, is a git repo if you want worktree spawning, and has accepted workspace trust; and the machine's memory, swap and prior OOM history.

Read the full file on GitHub · 177 lines

Files

What ships with it

8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 5d ago First seen · 177 lines · 81 tokens per session scan B 519a56bc42c8

Subscribe to this mod's changes

klimkit-remote-control is a skill published in the GitHub repository klimentij/klimkit (7 stars, last pushed 26d ago), licensed MIT. It adds 81 tokens to every session and 2,211 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

OpenClaw 网络优化助手

This skill should be used when the user asks to "优化网络", "配置 Clash 代理", "修复 GitHub 或 OpenAI 连接", "诊断 DNS 延迟", "设置 7897 代理端口", "网络变慢", or "输出 before/after 网络优化报告". It guides safe, evidence-first network diagnosis and reversible network changes with special protection for Clash, Mihomo, VPN, TUN, and developer CLI proxy…

rfdiosuao/AgentSkill · 102 tokens

maintain-splash-pages

The Lossless Group's pattern for repo-level splash pages — small Astro sites at /splash/ that ship to GitHub Pages on push to main, render the repo's changelog/ + context-v/ alongside curated marketing copy, and stay isolated from any package the repo also publishes. Use proactively whenever scaffolding a noteworthy…

lossless-group/lossless-agent-skills · 222 tokens

vercel-env-sync

Reconcile Vercel project environment variables against the local .env.local and your app's env schema. Use when the user wants to verify, diff, add, or remove env vars on Vercel so a deploy will boot — or when diagnosing 401 / [DISPATCHFAILED] symptoms that smell like missing secrets. Read-only by default; mutations…

dotlas/skills · 84 tokens

vercel-logs

Query and stream Vercel runtime/build logs for the production deployment via the CLI. Use when diagnosing errors, investigating 500s/401s, confirming a fix deployed cleanly, or watching a deploy land. Read-only — never deploys, rolls back, or mutates project state.

dotlas/skills · 64 tokens

infra-review

Review infrastructure-as-code and rendered deployment configuration without modifying it. Use before plan, apply, or deploy when targeting, replacement, state, availability, cost, ordering, or recovery blast radius may change.

yarlson/yarstack · 43 tokens

terraform-test

Comprehensive guide for writing and running Terraform tests. Use when creating test files (.tftest.hcl), writing test scenarios with run blocks, validating infrastructure behavior with assertions, mocking providers and data sources, testing module outputs and resource configurations, or troubleshooting Terraform test…

eric861129/SKILLS_All-in-one · 59 tokens