AReaL: Agent for Claude Code

.claude/agents/launcher-scheduler-expert.md

launcher-scheduler-expert is an agent for Claude Code from areal-project/AReaL. It costs 43 tokens per session (2,027 once invoked), scanned A, original, Apache-2.0.

An expert agent for launching distributed training or inference jobs and assigning cluster resources with Slurm, Ray, or Kubernetes.

In plain words
What is it for?
Use it when changing launcher or scheduler code, configuring cluster resources, troubleshooting deployments, or planning GPU allocation.
Why use it?
It helps diagnose failed launches, port conflicts, GPU allocation problems, and scheduling configuration issues across these systems.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

This is areal-project/AReaL's own configuration. It tells Claude Code how to work on AReaL 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 AReaL configures →

About the project

AReaL is an infrastructure system for training large language models with reinforcement learning, connecting model training to applications built around AI agents. Researchers and developers use it to train reasoning and agentic models through asynchronous workflows, and the catalogue add-ons support working with AReaL.

areal-project/AReaL · 5,748 stars · on GitHub · areal-ai.io

Reuse

Borrowing it

Nothing to install: this file belongs to areal-project/AReaL. 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/areal-project/AReaL/main/.claude/agents/launcher-scheduler-expert.md
Clone the repo
git clone --depth 1 https://github.com/areal-project/AReaL

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 launcher-scheduler-expert

README.md
[![agentmods](https://agentmods.dev/badge/agents/areal-project/areal/launcher-scheduler-expert/github.svg)](https://agentmods.dev/agents/areal-project/areal/launcher-scheduler-expert)
Your own site
<a href="https://agentmods.dev/agents/areal-project/areal/launcher-scheduler-expert"><img src="https://agentmods.dev/badge/agents/areal-project/areal/launcher-scheduler-expert/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.

agentmods 80×15 button for launcher-scheduler-expert

Your own site · 80×15
<a href="https://agentmods.dev/agents/areal-project/areal/launcher-scheduler-expert"><img src="https://agentmods.dev/badge/agents/areal-project/areal/launcher-scheduler-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,027 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00043 $0.02027
Opus 5 $0.00022 $0.01014
Sonnet 5 $0.00009 $0.00405
Haiku 4.5 $0.00004 $0.00203

Measured 11d ago against content hash 7577f7244f33, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

launcher-scheduler-expert 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 11d 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.

.claude/agents/launcher-scheduler-expert.md · 187 lines

How it starts

The opening of the file, as written. The whole thing — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Launcher & Scheduler Expert

You are an expert in distributed training cluster launching and resource scheduling, specializing in Slurm, Ray, and Kubernetes deployments for AReaL. Your role is to guide launcher/scheduler configuration, troubleshoot deployment issues, and ensure resource allocation correctness.

When to Activate

Use this agent when requested when:

  • Code modifications: User edits files in areal/infra/launcher/, areal/infra/rpc/, or areal/infra/scheduler/
  • Configuration changes: User modifies ClusterSpecConfig, SchedulerConfig, or related dataclasses
  • Deployment issues: User encounters job launch failures, port conflicts, GPU allocation errors
  • Resource planning: User needs guidance on cluster sizing, GPU allocation, or environment setup
  • Integration questions: User asks about launcher/scheduler interaction with engines/workflows

Core Concepts

Launcher vs. Scheduler

Component Responsibility Key Classes Config Source
Launcher Starts training/inference processes, manages process tree, passes environment variables LocalLauncher, SlurmLauncher, RayLauncher, SGLangServerWrapper, vLLMServerWrapper ClusterSpecConfig (cluster specification)
Scheduler Allocates GPU/port resources, manages worker lifecycle, performs health checks LocalScheduler, SlurmScheduler, RayScheduler SchedulerConfig (scheduling strategy)

Key Configuration Dataclasses

Located in areal/api/cli_args.py:

Read the full file on GitHub · 187 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. 11d ago First seen · 187 lines · 43 tokens per session scan A 7577f7244f33

Subscribe to this mod's changes

launcher-scheduler-expert is an agent published in the GitHub repository areal-project/AReaL (5,748 stars, last pushed today), licensed Apache-2.0. It adds 43 tokens to every session and 2,027 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 2026-08-30.

Related

Other agents, from other repositories

hybrid-cloud-architect

Design hybrid cloud infrastructure across AWS/Azure/GCP and OpenStack on-premises environments. Implement multi-cloud Terraform IaC, optimize costs, and manage hybrid connectivity. Handles auto-scaling, multi-region deployments, serverless architectures, and OpenStack private cloud. Use PROACTIVELY for hybrid cloud…

echoVic/blade-code · 77 tokens

kubernetes-architect

Design cloud-native infrastructure with Kubernetes at its core across AWS/Azure/GCP and hybrid environments. Implement GitOps workflows, OpenGitOps principles, and cloud-native patterns. Masters EKS, AKS, GKE, and self-managed clusters. Handles service mesh, observability, and progressive delivery. Use PROACTIVELY for…

echoVic/blade-code · 83 tokens

cloud-architect

Design AWS/Azure/GCP infrastructure, implement Terraform IaC, and optimize cloud costs. Handles auto-scaling, multi-region deployments, and serverless architectures. Use PROACTIVELY for cloud infrastructure, cost optimization, or migration planning.

echoVic/blade-code · 52 tokens

terraform-specialist

Write advanced Terraform modules, manage state files, and implement IaC best practices. Handles provider configurations, workspace management, and drift detection. Use PROACTIVELY for Terraform modules, state issues, or IaC automation.

echoVic/blade-code · 48 tokens

edge-proxy

Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for an edge and reverse-proxy tier (Cloudflare/Nginx/HAProxy/Traefik/Envoy/F5-BIG-IP/Citrix-ADC: origin discovery, admin planes, request smuggling, cache poisoning, WAF bypass).

ASCIT31/Dark-Moon · 0 tokens

gcp

Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for a Google Cloud Platform environment (IAM/service-accounts/impersonation/GCS/GCE/Functions/Run/GKE/SecretManager/BigQuery/CloudSQL/metadata).

ASCIT31/Dark-Moon · 52 tokens