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.
curl -O https://raw.githubusercontent.com/matthewdcage/llm-swarm-router/main/AGENTS.mdgit clone --depth 1 https://github.com/matthewdcage/llm-swarm-routerWrote 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/instructions/matthewdcage/llm-swarm-router/agents-md)<a href="https://agentmods.dev/instructions/matthewdcage/llm-swarm-router/agents-md"><img src="https://agentmods.dev/badge/instructions/matthewdcage/llm-swarm-router/agents-md/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/instructions/matthewdcage/llm-swarm-router/agents-md"><img src="https://agentmods.dev/badge/instructions/matthewdcage/llm-swarm-router/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.11066 | $0.11066 |
| Opus 5.5 | $0.04426 | $0.04426 |
| Sonnet 5.5 | $0.02213 | $0.02213 |
| Haiku 4.5 | $0.01107 | $0.01107 |
Grade B, and why
llm-swarm-router AGENTS.md 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 2mo 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.
Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- Do not commit `.cursor/plans/`, `.cursor/outreach/`, `.cursor/agents/`, `.cursor/coordinator/`, `tests/fixtures/coordinator-snapshot/`, `tests/fixtures/coordinator-monitor/`, `tests/test_coordinator_*.py`, coordinator- How it starts
The opening of the file, as written. The whole thing — 323 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent & developer guide
DOX rail
Full protocol: Child DOX Index below and per-folder AGENTS.md on each route (dox-agent plugin).
- Before editing: walk root → target path; read every
AGENTS.mdon the route (nearest doc controls local detail) - After meaningful edits: update the closest owning
AGENTS.md; refresh Child DOX Index when boundaries change - Child docs must not weaken parent DOX
Project overview
swarm-llm (netllm) is a mesh router for local LLM backends. Each host runs a lightweight agent that discovers oMLX (macOS), Ollama, LM Studio, and vLLM on localhost, finds sibling agents on the LAN via mDNS, and exposes dual API surfaces: OpenAI-compatible http://<host>:11400/v1 and Anthropic Messages API http://<host>:11400/v1/messages (with translation to local backends).
Tech stack: Python 3.11+, uv workspace monorepo, FastAPI agent, Typer CLI.
Architecture
| Package | Path | Role |
|---|---|---|
| netllm-core | packages/netllm-core/ |
Routing, health cache, config |
| netllm-sdk-openai | packages/netllm-sdk-openai/ |
OpenAI SDK upstream adapter |
| netllm-sdk-anthropic | packages/netllm-sdk-anthropic/ |
Anthropic SDK upstream adapter |
| netllm-discovery | packages/netllm-discovery/ |
Local scan, swarm registry, mDNS |
| netllm-agent | packages/netllm-agent/ |
FastAPI: /v1/*, /netllm/v1/*, /metrics |
| netllm-cli | packages/netllm-cli/ |
Typer CLI |
Honcho integration: docs/honcho-integration.md.
Repository layout
| Path | Purpose | DOX |
|---|---|---|
packages/ |
Python source of truth (uv workspace) | packages/AGENTS.md |
apps/ |
Native apps: macOS menubar today (apps/netllm-mac/) |
apps/AGENTS.md |
packaging/ |
Release builds per OS: packaging/README.md | packaging/AGENTS.md |
docs/ |
User install/troubleshoot/editor guides: docs/README.md | docs/AGENTS.md |
tests/ |
Cross-package integration tests | tests/AGENTS.md |
tests/conformance/ |
Conformance kits + ledgers (Axes A/B/D) | tests/conformance/AGENTS.md |
tests/extending/ |
Worked examples: a fixture registry entry, end to end | (tests rail) |
docs/extending/ |
Extension guides, program, evidence maps | docs/extending/AGENTS.md |
scripts/ |
CI, skill sync, install emulation | (root rail) |
Formula/ |
Homebrew formula | (root rail) |
archived/ |
Local deprecated/moved files (gitignored; not on remote) | (root rail) |
.agents/skills/ |
Canonical agent skills → sync via scripts/sync-agent-skills.sh |
.agents/AGENTS.md |
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.
- 2mo ago First seen · 323 lines · 11,066 tokens per session scan B 82cf2a578044
llm-swarm-router AGENTS.md is an instructions file published in the GitHub repository matthewdcage/llm-swarm-router (26 stars, last pushed 7d ago), licensed MIT. It adds 11,066 tokens to every session, about $0.0443 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
llm-swarm-router AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| llm-swarm-router AGENTS.mdmatthewdcage/llm-swarm-router | Instructions file | 26 | 7d ago | MIT | B | 11,066 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |