- Home
- Skills
- Agents & Orchestration
- multiplayer desync detective
Works with the AI Agents you already use
multiplayer desync detective
Diagnose Unreal Engine multiplayer desyncs by correlating server/client logs and auditing replication code.
$5.99
multiplayer desync detective
Example session with this skill installed
I'm seeing rubber-banding when players use the dash ability. It works fine in single player. Here are the logs from the listen server and one client. Check IBInfantryMovementComponent.cpp.
- Read your context and instructions
- Compiled the multiplayer desync detective
The logs show a clear prediction mismatch. The client triggers the dash immediately, but the server rejects the position update because the cooldown on the server had 12ms remaining when the move arrived.
Desync: Dash Rubber-banding (2024-05-22)
Verdict
Root Cause: Movement prediction mismatch caused by client-side clock drift and server-side cooldown enforcement.
Confidence: High.
Timeline
| Server Time | Client Time | Delta | Event |
|---|---|---|---|
| 00:42.100 | 00:42.088 | -12ms | Client: Ability_Dash triggered |
| 00:42.150 | 00:42.150 | 0ms | Server: Received Move, Cooldown active (0.012s) |
| 00:42.165 | 00:42.152 | +13ms | Server: LogCharacterMovement: ClientAdjustPosition |
| 00:42.200 | 00:42.190 | +10ms | Client: Position snapped back to pre-dash |
Root cause
The dash logic in IBInfantryMovementComponent.cpp starts the movement locally before the Server RPC confirms the cooldown has expired. Because the client clock is slightly ahead of the server's cooldown timer, the server processes the movement while bIsDashReady is still false, triggering a ClientAdjustPosition correction.
Evidence
[2024.05.22-14.42.150][402] LogNet: Received RPC: ServerMove[2024.05.22-14.42.165][403] LogCharacterMovement: ClientAdjustPosition: Bouncing client due to collision or invalid move
Fix
Move the dash cooldown check into the FSavedMove_IBInfantry flow to ensure the server and client evaluate the cooldown at the same relative timestamp.
- Update
IBInfantryMovementComponent.cppto include cooldown state in the saved move. - Ensure
CanAttemptDash()accounts fornet.PktLag.
Verification
- Run PIE with 2 players.
- Set
net.PktLag 150. - Enable
p.NetShowCorrections 1. - Trigger dash repeatedly; logs should no longer show
ClientAdjustPositionduring cooldown transitions.
Next steps
- Verify if
bIsDashReadyis replicated withCOND_OwnerOnly. - Review the
Infantry/IBInfantryMovementComponentfor timestamp synchronization.
Connects securely to your tools. The creator never sees your data.
About this skill
In a server-authoritative environment, even a minor disagreement between machines can break gameplay. This skill identifies why your server and clients are out of sync, whether it is rubber-banding, hits failing to register, or state mismatches during late joins. It moves past symptom-guessing to find the exact frame where machines diverged.
What it does
- Log synchronization aligns timestamps and frames across server and client logs to build a unified timeline.
- Root cause isolation identifies the first point of divergence before symptoms become visible.
- Pattern matching scans for specific
LogNetandLogNetPackageMapwarnings like missing owning connections or unresolvable objects. - Replication auditing checks property conditions, RPC specifiers, and authority guards in code or Blueprints.
- Movement analysis diagnoses prediction errors and corrections in character movement components.
How it works
- Input logs from a single session including the server and at least one client.
- Build a timeline of gameplay events like spawning, damage, and ability activation with calculated deltas.
- Scan for suspects such as client-side mutation without authority, dropped unreliable RPCs, or OnRep failures.
- Inspect code for the specific replicated properties and authority checks that explain the logs.
Frameworks & tools
This skill is optimized for Unreal Engine multiplayer development. It analyzes C++ gameplay code, Blueprint nodes, and Unreal logging systems including LogNet, LogRep, and LogCharacterMovement.
Why this beats prompting it yourself
Generic LLMs struggle to correlate logs from two different machines or understand Unreal's specific authority and replication rules. This skill follows a strict technical methodology to find the "quiet" early divergence rather than reacting to the "dramatic" late symptom.
Use cases
- Debugging rubber-banding or movement corrections during high-latency scenarios.
- Fixing "ghost" abilities that fire on the client but do nothing on the server.
- Resolving state mismatches for players who join a match in progress.
- Identifying why a specific property fails to update on remote clients.
Known limitations
Requires logs with timestamps from both the server and client in the same session. Does not fix build version mismatches where machines are running different code.
Requires logs with timestamps from both server and client. Does not resolve issues caused by version mismatches between machines. Project-specific fixes follow strict authority rules.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 9 days ago