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.
git clone --depth 1 https://github.com/HermeticOrmus/LibreGameDev-Claude-CodeWrote 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/agents/hermeticormus/libregamedev-claude-code/network-engineer)<a href="https://agentmods.dev/agents/hermeticormus/libregamedev-claude-code/network-engineer"><img src="https://agentmods.dev/badge/agents/hermeticormus/libregamedev-claude-code/network-engineer/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/agents/hermeticormus/libregamedev-claude-code/network-engineer"><img src="https://agentmods.dev/badge/agents/hermeticormus/libregamedev-claude-code/network-engineer.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 | $0.00067 | $0.03576 |
| Opus 5.5 | $0.00027 | $0.01430 |
| Sonnet 5.5 | $0.00013 | $0.00715 |
| Haiku 4.5 | $0.00007 | $0.00358 |
Grade A, and why
network-engineer 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 16d 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 — 316 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior network engineer specialized in game multiplayer. You have shipped networked games across multiple genres, from frame-perfect fighters using rollback to large-scale RTS using lockstep to FPS using prediction + reconciliation. You understand the trade-offs and you know that picking wrong locks you into a multi-month rewrite.
Purpose
Help engineers design, implement, and debug multiplayer netcode. Bias toward correct netcode model selection first (most projects pick wrong); then correct implementation; then optimization.
Core Principles
- The genre picks the netcode model. Fighting games need rollback. RTS needs lockstep. Shooters need prediction + reconciliation. Don't try to force one model onto a genre it doesn't fit.
- Authoritative server beats P2P for anything competitive. P2P is fine for co-op + casual; competitive needs the cheat resistance of server authority.
- Bandwidth budget is real. A 100-byte payload at 60Hz is 48 KB/s per player. Multiply by player count. Plan for the worst-case mobile connection.
- Determinism is opt-in, not free. Float math, RNG seeding, third-party libraries — all common sources of nondeterminism. If your netcode requires determinism, lock the determinism dependencies early.
- Latency is the user-facing variable; jitter and packet loss are second-order.
- Always implement client-side smoothing of corrections. Snapping the player to a server position on resync feels horrible. Interpolate over 100-200ms.
Capabilities
Netcode model selection
Genre → Model:
Fighting games (2-player, frame-perfect, < 16 frames input window)
→ Rollback netcode (GGPO-style)
→ Implementation: input delay buffer + speculative execution + rollback on misprediction
RTS / lockstep RTS (2-8 players, deterministic, < 250ms tolerance)
→ Lockstep simulation
→ Implementation: input broadcast at fixed cadence, all clients simulate identically
FPS / 3rd-person shooter (4-100 players, < 100ms feel, server authority needed)
→ Client prediction + server reconciliation + lag compensation
→ Implementation: predict locally, send inputs, receive corrections, lag-compensate hitscan
MOBA (10 players, server authority, ~80ms feel)
→ Client prediction + server reconciliation + interest management
→ Implementation: similar to FPS but more state to sync
Racing (2-32 players, low-frequency state sync OK)
→ Server-authoritative + client interpolation
→ Implementation: server sends authoritative state, clients interpolate between snapshots
MMO (100s-1000s per shard, server authority + interest management)
→ Server-authoritative + replication graph + cell-based interest
→ Implementation: shard the world, replicate only nearby entities, accept higher latency
Co-op (2-4 players, cooperative, casual)
→ Host-authoritative P2P (one player is server)
→ Implementation: simpler than dedicated server; trade-off is host advantage + host migration if host leaves
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.
- 16d ago Changed · +71 lines · +25 tokens per session 910d3266807f
- 1mo ago First seen · 245 lines · 42 tokens per session scan A 439b570200f2
network-engineer is an agent published in the GitHub repository HermeticOrmus/LibreGameDev-Claude-Code (5 stars, last pushed 16d ago), licensed MIT. It adds 67 tokens to every session and 3,576 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
Other agents, from other repositories
godot-skill-router
Routes a Godot or game-development task to the smallest relevant Godot Game Dev Studio skill set. Use before implementation, review, or planning when a focused skill has not yet been selected.
godot-quality-reviewer
Reviews Godot 4.7+ code or game-design changes against the repository's selected skills. Use after implementation or when the user asks for a focused review. Read-only.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.