Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/bam-bam-2/solo-skillsnpx agentmods add skills/bam-bam-2/solo-skills/remote-offloadWrote 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/skills/bam-bam-2/solo-skills/remote-offload)<a href="https://agentmods.dev/skills/bam-bam-2/solo-skills/remote-offload"><img src="https://agentmods.dev/badge/skills/bam-bam-2/solo-skills/remote-offload/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bam-bam-2/solo-skills/remote-offload"><img src="https://agentmods.dev/badge/skills/bam-bam-2/solo-skills/remote-offload.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00061 | $0.02000 |
| Opus 5 | $0.00030 | $0.01000 |
| Sonnet 5 | $0.00012 | $0.00400 |
| Haiku 4.5 | $0.00006 | $0.00200 |
Grade A, and why
remote-offload 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 7d 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.
원격 머신 오프로딩
이 스킬이 시스템에 하는 일 (설치 전 확인)
~/.ssh/config에 등록된 원격 호스트로 SSH 접속해 명령을 실행합니다.- 원격이 응답하지 않으면 로컬로 대체 실행하지 않고 중단합니다. 의도된 동작입니다.
- 새 파일이나 상주 프로세스를 만들지 않습니다.
--bg사용 시 원격에 로그 파일(/tmp/offload-*.log) 하나가 남습니다.
바로 쓰는 스크립트
scripts/offload.sh — 무거운 작업을 원격 기기로 넘깁니다.
./scripts/offload.sh "python heavy_scraper.py"
./scripts/offload.sh --dir ~/Projects/app "npm run build"
./scripts/offload.sh --bg "ffmpeg -i in.mov -c:v libx264 out.mp4" # 백그라운드
원격이 죽어 있으면 로컬로 폴백하지 않고 그냥 멈춥니다. 메모리 부족으로 노트북이 죽는 걸 막기 위해서입니다.
왜
| Aside 돌아가는 기기 | 오프로딩 대상 | |
|---|---|---|
| 기기 | MacBook Air, Apple M1 | Mac mini, Apple M4 |
| 램 | 8 GB (상시 부족, 압축 2.6GB / 여유 70MB 수준) | 16 GB |
| 절전 | 사용자가 덮으면 멈춤 | pmset sleep 0 — 안 잠듦 |
| 디스크 여유 | 200 GB | 54 GB |
맥북에어는 램이 모자라 크로미움이 백그라운드 탭을 강제 종료한다. 로컬에서 무거운 걸 돌리면 사용자의 탭이 죽는다. 무거운 건 원격 머신로 넘기는 게 기본값이다.
접속
ssh remote-host # Tailscale (REMOTE_IP) — 기본으로 이걸 쓴다
ssh remote-host-lan # 같은 LAN일 때 (remote-host.local)
비대화형에서는 -o BatchMode=yes -o ConnectTimeout=10을 붙인다.
PATH가 필요한 명령(brew 설치물 등)은 ssh remote-host 'bash -lc "..."'로 로그인 셸을 태운다.
작업공간 규약
~/Projects/<프로젝트>/offload/
├── run.sh 러너 (아래 참조)
├── jobs/ 실행할 스크립트를 여기로 scp
├── out/ 산출물 (이미지, csv, json 등)
├── logs/ <JOB_ID>.log / .pid / .cmd.sh
└── node_modules/ playwright 설치됨
기존 프로젝트 작업은 ~/Projects/<repo>에서 그대로 하고, 일회성 오프로딩만 여기를 쓴다.
판단 기준
넘긴다
- 헤드리스 스크래핑, 특히 여러 페이지 순회 (로그인 불필요한 공개 페이지)
- ffmpeg 인코딩, GIF 생성, 이미지 대량 변환
npm install,next build, 대형 저장소 clone- 대량 API 수집 (수백~수천 건 페이지네이션)
- 수십 MB 이상 파일 파싱/집계
- 30초 넘게 걸릴 것 같은 CPU 작업 전부
- 긴 문서/영상 스크립트 읽고 요약 (아래
claude -p레시피)
안 넘긴다
- 사용자의 로그인 세션·쿠키가 필요한 브라우징 → Aside 브라우저에서 해야 함
- 사용자가 화면으로 확인해야 하는 UI 조작
- 몇 초 안에 끝나는 단순 조회 (ssh 왕복이 더 느림)
- 사용자 개인 파일이 맥북에어에만 있는 경우 (전송 비용 판단 후 결정)
표준 패턴: 스크립트를 보내서 실행
ssh + bash -lc + heredoc을 중첩하면 따옴표가 반드시 깨진다. 로컬에 파일로 쓰고 scp로 보낸 뒤 실행한다.
What ships with it
1 file 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.
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.
- 7d ago Changed · +11 lines 5314bb70d794
- 11d ago First seen · 159 lines · 61 tokens per session scan A 254661604721
remote-offload is a skill published in the GitHub repository bam-bam-2/solo-skills (362 stars, last pushed 7d ago), licensed MIT. It adds 61 tokens to every session and 2,000 once invoked, about $0.0003 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 skills, from other repositories
calcom-access
Access Cal.com API via CLI with 1Password API key. Use when user wants to list bookings, create event types, manage schedules, or.
n8n-workflows
Use when n8n workflow automation — nodes, triggers, expressions, credentials, webhooks, error handling. Use when working with n8n workflows.
pipedream-workflows
Use when pipedream serverless workflows — triggers, code steps, pre-built actions, data stores, HTTP. Use when working with pipedream workflows.
flowise-builder
Use when flowise visual LLM workflow builder — drag-drop chatflows, API endpoints, document loaders, tools. Use when working with flowise builder.
postbridge-social-manager
Use when multi-platform social media posting, scheduling, analytics, and media management via PostBridge API for TikTok, Instagram, X, LinkedIn, and Facebook. Use when working with postbridge social manager.
telegram-userbot
Use when full MTProto control of Telegram account via Telethon. DM, Voice Note, Call, Video Call, Group/Channel management, member scraping, bot cloning, outreach automation, broadcast, CRM tracking, content reposting, scheduled messaging, webhook triggers. Use for all Telegram automation as a real user (not bot API).