agentleFS
rodrigo-guerraarodrigo-guerraa/roblox-hub/skills/vendored/roblox-networking/SKILL.md

roblox-networking

Skill2 starsChanged 30 days agoMITView on GitHub

Roblox client-server networking: RemoteEvents, RemoteFunctions, UnreliableRemoteEvents, bindables, argument validation, rate limits, server authority, payload budgets, hit validation with GetServerTimeNow, TextChatService. Use when adding, changing, debugging or auditing anything sent between client and server: remote validation, cooldowns, laggy or dropped events, desync, payload size, exploit-proofing remotes.

6 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use roblox-physics, roblox-audio and 4 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • szlaySkill

    roblox-physics

    Use when building Roblox vehicles, ragdolls, projectiles, elevators, constraints, forces, or other physics-driven gameplay.

    Also in 5 projects that use this

  • szlaySkill

    roblox-audio

    Use when implementing Roblox audio playback, spatial sound, music, sound effects, SoundGroups, or dynamic audio effects.

    Also in 4 projects that use this

  • sentinelcoreSkill

    roblox-performance

    Use when optimizing a Roblox game for better frame rates, reducing lag, improving server or client performance, diagnosing FPS drops, handling large worlds, or when asked about streaming, draw calls, object pooling, LOD, MicroProfiler, or expensive loop operations.

    Also in 3 projects that use this

  • +3 more from szlay

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every client argument is attacker-controlled: validate type, size, ownership, state, distance and cooldown on the server.

    line 16
  2. Distinctive:

    Look up prices, damage, rewards and permissions from server-owned definitions.

    line 17
  3. Very specific:

    Check NaN/infinity first (x ~= x, math.abs(x) == math.huge).

    line 25

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Roblox networking

Where this conflicts with the roblox-lessons skill, roblox-lessons wins.

When to use

Adding a remote, handling untrusted input, adding cooldowns, assigning authority, or reviewing remote security. Broader exploit audits: roblox-security.

Quick reference

  • Every client argument is attacker-controlled: validate type, size, ownership, state, distance and cooldown on the server. Exploits with bad arguments are rejected silently.
  • Look up prices, damage, rewards and permissions from server-owned definitions; the client asks for "buy item Y", never "subtract 20 coins".
  • Choose the authority model first. Server Authority (Workspace.AuthorityMode = Server plus its settings bundle) uses client prediction and rollback; it is not SetNetworkOwner. Simulation input there uses InputAction + BindToSimulation(), not a remote.
  • RemoteEvent for most requests; keep RemoteFunction calls short and bounded.
  • RemoteEvent is reliable but not ordered against property/attribute replication; use one explicit channel or version state. UnreliableRemoteEvent only for replaceable data (droppable, 1000-byte cap).
  • Engine ceiling: about 500 FireServer requests/s per client, pooled across remotes of the same type. A ceiling, not a target.
  • Budget rate x bytes x recipients; snapshot on join, diffs after.
  • Hit timestamps are untrusted context, never proof; bound freshness and validate against server history (full.md §3a).
  • Typed schemas/IDLs do not replace security checks. Pin generators and verify generated output in CI.
  • Check NaN/infinity first (x ~= x, math.abs(x) == math.huge); NaN defeats </>. utf8.len catches malformed UTF-8 that would fail a DataStore save.
  • Serialization: functions arrive nil, metatables stripped, mixed keys mangled, nil truncates arrays, tables are copies. Validate field by field; share state via server-owned snapshots or ids.
  • Rate limits protect the server; validation still rejects invalid requests. Bindables are same-side only.
  • Record suspicion with thresholds; never punish one malformed packet.
  • ProximityPrompt/ClickDetector/DragDetector triggers are remote-equivalent input: validate them the same way.

Initial sync (from roblox-lessons data.md)

A queued RemoteEvent payload is delivered only to the first client listener. If several client scripts need an on-join push, each fires its own Request* remote on startup; the server handler only re-sends, never mutates. A loading screen waits for the list of on-join events, re-requests any it missed, and times out with a warning naming them.

Validation, throttling, budgets, chat and mock patterns: references/full.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.