givemeanode-agent-compute

givemeanode-agent-compute is a skill for Claude Code, Codex from synthetic-sciences/openscience. It costs 46 tokens per session (1,162 once invoked), scanned A, original, Apache-2.0.

A control interface for GiveMeANode, a service that provides rented GPU computers, batch jobs, storage, and temporary work environments. It checks current limits, prices, resources, and job ownership before actions.

In plain words
What is it for?
Use it to inspect GPU availability and billing, run batch workloads, manage storage, and review or clean up jobs with approval for paid or destructive actions.
Why use it?
It helps prevent unexpected spending, unsafe deletion, and confusion in shared workspaces. It also supports recovery when remote work needs to continue after interruption.

Skill for Claude CodeCodex

About the project

synthetic-sciences/openscience is an AI workbench that carries out scientific research by reading papers, forming hypotheses, writing and running code, conducting experiments, analyzing results, and preparing reports. Researchers use it for work in machine learning, biology, physics, and chemistry with remote or local models. Catalogue add-ons extend its scientific workflows through skills and instructions.

synthetic-sciences/openscience · 3,432 stars · on GitHub

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/synthetic-sciences/openscience/givemeanode
Any agent
npx skills add synthetic-sciences/openscience --skill givemeanode
Clone the repo
git clone --depth 1 https://github.com/synthetic-sciences/openscience

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 givemeanode-agent-compute

README.md
[![agentmods](https://agentmods.dev/badge/skills/synthetic-sciences/openscience/givemeanode.svg)](https://agentmods.dev/skills/synthetic-sciences/openscience/givemeanode)
Your own site
<a href="https://agentmods.dev/skills/synthetic-sciences/openscience/givemeanode"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/givemeanode.svg" alt="Measured on agentmods" height="20"></a>
Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,162 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00046 $0.01162
Opus 5 $0.00023 $0.00581
Sonnet 5 $0.00009 $0.00232
Haiku 4.5 $0.00005 $0.00116

Measured yesterday against content hash 0e2af7e55e3e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

givemeanode-agent-compute 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 yesterday.

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.

backend/cli/skills/cloud-compute/givemeanode/SKILL.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GiveMeANode through MCP

Use the configured givemeanode MCP connector. Do not install or invoke gman, call the HTTP API directly, or ask for a service token when the MCP tools are available. The connector owns OAuth; OpenScience's MCP permission boundary owns each tool invocation.

GiveMeANode is not currently a native compute_job target. Use its discovered MCP tools directly, and do not claim that a remote node or batch job is tracked, recoverable, or releasable through OpenScience's JobBroker.

Start with live state

  • Read the available MCP tools and the gman://docs/llms.txt resource before relying on remembered arguments. The service contract is versioned and /preview may evolve.
  • Read list_limits, get_billing, relevant nodes/jobs, and the current published rate before proposing paid work. Never hard-code prices, availability, queue time, images, or limits. If role-scoped billing inspection is refused, surface that gap instead of assuming funds or headroom.
  • In a shared workspace, pass a stable mission and inspect yours, created_by, and same_mission before stopping or deleting anything.

The exact locked rate and job cost ceiling arrive in the create/submit response, not from a separate quote operation. Before that mutation, show the current published estimate, the requested maximum duration, and the applicable workspace or organization cap. If the user has not authorized work with a price that is only locked on submission, do not submit it. Preserve the returned rate and cost ceiling in the handoff.

Choose the right primitive

  • Use an interactive node when work needs a persistent encrypted disk or repeated commands. Stopping parks the disk and stops billing; processes do not survive a stop.
  • Use a batch job for unattended, queueable work with declared outputs, checkpoints, or sweeps. Prefer this over holding an idle node.
  • Use a detached command for node work that may outlive one tool call. Retain its command ID and page logs by offset.
  • Use rollout sandboxes for short-lived fork/snapshot workloads, not as a substitute for persistent training storage.

Read the full file on GitHub · 63 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. yesterday First seen · 63 lines · 46 tokens per session scan A 0e2af7e55e3e

Subscribe to this mod's changes

givemeanode-agent-compute is a skill published in the GitHub repository synthetic-sciences/openscience (3,432 stars, last pushed yesterday), licensed Apache-2.0. It adds 46 tokens to every session and 1,162 once invoked, about $0.0002 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 2026-09-03.