agentmods.dev

llm-swarm-router: Skill for Claude Code

.agents/skills/netllm-swarm/SKILL.md

netllm-swarm is a skill for Claude Code from matthewdcage/llm-swarm-router. It costs 100 tokens per session (1,871 once invoked), scanned A, original, MIT.

A guide for connecting several computers on the same local network into one shared pool of language-model servers. It covers peer discovery, request routing, and optional gateway settings.

In plain words
What is it for?
Use it to create a netllm swarm, connect macOS, Linux, or Windows machines, discover peers with mDNS or network scanning, configure a gateway, and check that the mesh works.
Why use it?
It removes the manual work of making machines find one another and configuring clients to send requests to the shared pool.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: installed under .agents/ (shared by several agents); mentions Codex.

This is matthewdcage/llm-swarm-router's own configuration. It tells Claude Code how to work on llm-swarm-router 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 llm-swarm-router configures →

not rated 25repo 9d ago A scan SkillSpector: warn 100 tokens original MIT
View source ↗ matthewdcage/llm-swarm-router
Reuse

Borrowing it

Nothing to install: this file belongs to matthewdcage/llm-swarm-router. 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/matthewdcage/llm-swarm-router/main/.agents/skills/netllm-swarm/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/matthewdcage/llm-swarm-router

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 netllm-swarm

README.md
[![agentmods](https://agentmods.dev/badge/skills/matthewdcage/llm-swarm-router/netllm-swarm/github.svg)](https://agentmods.dev/skills/matthewdcage/llm-swarm-router/netllm-swarm)
Your own site
<a href="https://agentmods.dev/skills/matthewdcage/llm-swarm-router/netllm-swarm"><img src="https://agentmods.dev/badge/skills/matthewdcage/llm-swarm-router/netllm-swarm/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 netllm-swarm

Your own site · 80×15
<a href="https://agentmods.dev/skills/matthewdcage/llm-swarm-router/netllm-swarm"><img src="https://agentmods.dev/badge/skills/matthewdcage/llm-swarm-router/netllm-swarm.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 100 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,871 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Anti-Refusal · line 169
    Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
    Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
How audits are shown
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 $0.00100 $0.01871
Opus 5 $0.00050 $0.00936
Sonnet 5 $0.00020 $0.00374
Haiku 4.5 $0.00010 $0.00187

Measured 11d ago against content hash 089f919a8f3b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

netllm-swarm scanned grade A 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 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://127.0.0.1:11400/metrics | rg netllm_requests_total
.agents/skills/netllm-swarm/SKILL.md · 171 lines

How it starts

The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.

netllm swarm

When to use this skill

  • Multiple machines should share one logical model pool
  • User says "LAN swarm", "multi-machine netllm", "find peers", or /netllm-swarm
  • mDNS or static peer configuration needed

Prerequisites

  • netllm installed on each machine (netllm-setup skill or ./netllm init)
  • Same LAN/VLAN (guest Wi-Fi often blocks mDNS)
  • Local inference on at least one host (optional on workers if they only gateway)

Workflow (DMG / trusted home LAN — default)

  1. Each machine: install app → welcome Listen on LAN (or Settings LAN toggle) → restart agent
  2. Clients: export OPENAI_BASE_URL=http://<any-LAN-IP>:11400/v1 (Honcho, Cursor, etc.)
  3. Verify: ./netllm peers — agents find each other via mDNS / subnet scan (no token)

Open swarm: empty swarm.cluster_token is intentional on trusted LANs.

Workflow (CLI — same open default)

  1. Each machine:

    ./netllm init --swarm    # upgrades existing config; no --force
./netllm serve

    Binds 0.0.0.0:11400, selects local_spillover, enables subnet_scan — no token.

  2. Verify the mesh

    ./netllm peers          # both machines listed; flags unreachable loopback binds
./netllm models --lan
./netllm status

Workflow (manual — existing configs)

  1. Bind for LAN on each host

    ./netllm serve --host 0.0.0.0

    Packaged installs (systemd, Windows service, macOS app) may already bind via config, use ./netllm start / ./netllm restart after changing listen address. Confirm agent.advertise = true in ~/.config/netllm/config.toml (default in config.example.toml).

  2. Firewall, allow inbound TCP 11400 and UDP 5353 (mDNS) on each host — ./netllm doctor prints per-platform commands.

  3. Security on untrusted LANs (optional): Settings Require cluster token, or ./netllm init --swarm --secure, or ./netllm swarm-token --create. Then on other machines: ./netllm join http://<leader-ip>:11400 --token <token>. Heartbeats require Bearer token when set; open swarms reject join-with-token.

Read the full file on GitHub · 171 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 · 171 lines · 100 tokens per session scan A 089f919a8f3b

Subscribe to this mod's changes

netllm-swarm is a skill published in the GitHub repository matthewdcage/llm-swarm-router (25 stars, last pushed 9d ago), licensed MIT. It adds 100 tokens to every session and 1,871 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

CloudData and AIDevOps and cloudLLManthropic-apicursorllama-cpplm-studiolocal-llm
Related

Other skills, from other repositories

torchtitan

Pretrain LLMs at scale with PyTorch 4D parallelism.

NousResearch/hermes-agent · 20 tokens

modal

Serverless GPU cloud for ML jobs and model APIs.

NousResearch/hermes-agent · 13 tokens

tensorrt-llm

High-throughput LLM inference on NVIDIA GPUs.

NousResearch/hermes-agent · 18 tokens

lambda-labs

On-demand GPU cloud instances for ML training.

NousResearch/hermes-agent · 13 tokens

ultralytics-platform

This skill should be used when user asks to "upload my model to Ultralytics Platform", "push this run to the platform", "upload a dataset to platform", "download a dataset from platform", "search platform datasets", "start cloud training", "train on platform GPUs", "export a model on platform", "deploy a model…

fcakyon/claude-codex-settings · 112 tokens

managed-model-endpoints

Register a model service in the managed family — a local model server container the daemon starts/stops on demand, or a remote upstream model API (https). Read the runbook, allocate a port (local only), compose idempotent start/stop scripts (local only), register once. Load when the user wants a model service…

UnicomAI/wanwu · 83 tokens