Mimics Mod for R.E.P.O

Mimics adds a voice-mimic mechanic to monsters in R.E.P.O. They can replay random phrases recorded during gameplay, making it harder to trust what you hear.

📖 Description

Think that was your teammate calling for help? Think again. Mimics brings a Skinwalker-style mechanic into R.E.P.O by allowing monsters to mimic player voices using random recorded phrases.

🎯 Who Needs This Mod

Both Host and Clients need to install this mod to work.

🧩 Compatibility

Works with R.E.P.O v0.5.0.

REPO_SteamNetworking_Lib 0.1.2 is broken on v0.5.0 (it still hooks the removed SteamManager). Mimics ships a built-in compatibility fix that re-attaches the library to the new SteamLobbyManager, so other mods that depend on the library keep working too while Mimics is installed. The Undefined target method ... SteamManagerPatches error at startup comes from the library itself and is harmless. The fix turns itself off once an updated library hooks the game on its own.

All players in a lobby need the same Mimics version.

This mod is incompatible with mods:

1. Disable Noise Reduction by SVRZ, as it relies on the game's built-in noise reduction to capture audio and send chunks to other players correctly.

✨ Features

• Monsters mimic player voices recorded during the current session
• In multiplayer, all players hear the same mimic clip on the same enemy at the same time
• Voice can be played back with a random pitch or alien filter — the host decides which, so all players hear the same effect
• Smart director — the host picks the voice, the enemy and the listeners by context instead of at random (see below)
• Event reactions — mimics react to deaths, damage, enemies spotting players, broken or grabbed valuables, and the Head Grabber carrying a player's head
• Microphone setup wizard on the main menu tunes how loud you need to speak to be recorded
• Debug menu (F8) with a live mic meter, recordings manager, director decisions and event feed
• Custom audio support — drop your own .mp3 or .wav files into the custom-audio folder to add more sounds to the mimic pool
• Mod authors can register custom clips at runtime via the public MimicsAPI
• Built for long sessions: per-player memory limit, and players who join mid-session don't break the voice pool

🧠 Smart Director

The host scores every voice × enemy × listener combination and plays the most convincing one (with a bit of randomness):

• A voice of someone the listener can see, or who stands right next to them, is avoided — that would give the mimic away
• Voices of far away or dead players are preferred
• Enemies hidden behind walls or out of view are preferred over ones in plain sight
• Lonely listeners are targeted more often
• Recently and frequently played clips are used less
• Your own voice is never played to you if Hear Yourself? is off

Event reactions

Event	What happens
Player death	8–20 s later, an enemy near the body calls out with the dead player's voice
Player hurt	Shortly after, teammates nearby hear the injured player's voice from an enemy
Enemy spotted player	That enemy tries to lure the player with another teammate's voice
Valuable broken	An enemy nearby reacts with the voice of whoever held it
Valuable picked up	Sometimes a hidden enemy nearby talks to the player who grabbed loot worth $200+
Head Grabber carries head	The carried head calls out with its owner's voice and screams when the grabber attacks — the dead owner watches through it

If no enemy is close enough when an event happens, it waits a few seconds for one to come near. Every event can be turned off in the Host Only settings, and Smart Director = false switches back to the classic random mode.

🎙️ Microphone Setup Wizard

On first launch (and after updates that change it) a short wizard opens on the main menu:

1. Checks the microphone selected in the game's audio settings
2. Measures your room's background noise (3 s of silence)
3. Measures your normal speaking level (5 s of talking)
4. Recommends a Voice Detection Sensitivity and lets you test it live

Reopen it any time by turning on Show Setup Wizard in the mod settings.

🧪 Debug Menu

Press F8 in a level, the shop or the truck lobby. Shift+F8 gives the mouse back to the game while the window stays visible. Rebind the key by pressing a new one in the Settings tab.

• Status strip — live mic level with noise floor and detection threshold markers, current playbacks
• Recordings — every cached voice and custom clip with waveform, play count and age: listen locally, play on the nearest enemy, pin (kept from auto-cleanup) or delete
• Director — pending events, recent decisions with a per-listener score breakdown, the event feed, and buttons that simulate events for solo testing
• Mobs / Players / Network / Overlays / Settings — enemy list, player state and per-player mimic volume, transmissions and sound readiness, in-world overlays, all settings

📦 Installation via r2modman

1. Install r2modman from https://github.com/ebkr/r2modmanPlus/releases
2. Open r2modman and select R.E.P.O
3. Create or choose a profile
4. Search for Mimics by ToxesFoxes
5. Install the mod

Dependencies are pulled automatically:

• BepInEx-BepInExPack-5.4.2305
• REPOModding-REPOConfig-1.2.7
• REPOModding-MenuLib-2.6.0
• Rune580-REPO_SteamNetworking_Lib-0.1.2

⚙️ Mod Settings

Settings are stored in the BepInEx config file:

<REPO_PROFILE>/BepInEx/config/TFS_Mimics.cfg

General

Setting	Default	Acceptable Values	Description
Volume	30	0-100	Mimic playback volume in percent.
Hear Yourself?	true	true / false	If false, your own voice is never played back to you; other players still hear it.
Voice Detection Sensitivity	85	0-100	How easily your voice starts a recording. Higher catches quieter speech but also more background noise. Set by the setup wizard.
Show Setup Wizard	false	true / false	Opens the microphone setup wizard on the main menu. Turns itself off after Save / Skip.
Max Clips In Memory Per Player	40	5-500	Recordings per player kept loaded during a session. Oldest unpinned ones are dropped from memory, not from disk.
Persist Audio Cache	false	true / false	Enables saving received mimic clips to disk between sessions.
Persist Max Files Per Player	100	1-5000	Max saved recordings per player folder. Pinned recordings don't count.
Normalize Target	85	0-100	Peak normalization target for voice and custom audio (0 = off, 100 = 0 dBFS).

Host Only

These settings only take effect on the player who is host. Clients can leave them at defaults.

Setting	Default	Acceptable Values	Description
Playback Near Radius	15	5-100	Preferred radius around players for selecting a mimic target enemy.
MinDelay	10	1-300	Minimum seconds between mimic playback cycles.
MaxDelay	20	5-600	Maximum seconds between mimic playback cycles.
Playback Voice Filters Enabled	true	true / false	If true, the host randomly applies voice filter to playback. All players hear the same effect.
Smart Director	true	true / false	Context-aware choice of voice, enemy and listeners. false = classic random selection.
Event: Player Death	true	true / false	The dead player's voice calls out near the body.
Event: Player Hurt	true	true / false	Teammates nearby hear the injured player's voice.
Event: Enemy Spotted Player	true	true / false	The enemy lures the spotted player with another teammate's voice.
Event: Valuable Broken	true	true / false	An enemy reacts with the voice of whoever held the broken valuable.
Event: Head Grabber Talks	true	true / false	A head carried by the Head Grabber talks with its owner's voice.
Event: Valuable Picked Up	true	true / false	Sometimes a hidden enemy talks to the player who grabbed a valuable worth $200+.

Filter

• Filter Enabled? (default: false) — enables per-enemy mimic filter.

Experimental

• Sampling Rate (default: 48000, range: 16000-48000) — microphone and processing sample rate.

Debug

• Menu Keybind (default: F8) — key that opens the Mimics debug menu (any Unity KeyCode name, or rebind by key press in the menu).
• Verbose Logging (default: false) — enables detailed debug logs.

📁 Custom Audio

Drop your own sound files here to add them to the mimic pool:

<REPO_PROFILE>/BepInEx/plugins/ToxesFoxes-Mimics/custom-audio/

Supported formats: .mp3, .wav.

If persistence is enabled, recorded clips are saved here:

<REPO_PROFILE>/BepInEx/plugins/ToxesFoxes-Mimics/audio-cache/
    players.json
    pinned.txt
    <player_id>/audio_<player_id>_<guid>.bin

🧭 Roadmap

Done: context-aware playback tied to gameplay events (enemy sees player, player damage and death, broken items) — see Smart Director.

Ideas for later:

• Sending a few recent clips to players who join mid-session
• More enemy-specific behaviours, like the Head Grabber talking head

📝 Requirements

• R.E.P.O v0.5.0 (Steam)
• BepInEx pack for R.E.P.O
• Rune580-REPO_SteamNetworking_Lib
• Windows OS

⚠️ Disclaimer

• This mod is not affiliated with or endorsed by the developers of R.E.P.O
• Use at your own risk
• Always back up your saves before using any mods

📞 Support

• 🐛 Report Issues
• 📧 Contact: [email protected]

---

🔧 Manual Installation

1. Install BepInEx for R.E.P.O
2. Download the latest release from:
   • Thunderstore: https://thunderstore.io/c/repo/p/ToxesFoxes/Mimics/
   • GitHub: https://github.com/ToxesFoxes/Repo-Mimics/releases
3. Copy TFS_Mimics.dll into your plugins folder:

<REPO_PROFILE>/BepInEx/plugins/ToxesFoxes-Mimics/TFS_Mimics.dll

1. Launch the game

🔍 Differences From Original Mimic

This mod is a full rework focused on compatibility with newer R.E.P.O versions and long-term stability.

• Reworked enemy discovery for newer game versions instead of relying on a fixed scene path.
• New audio transmission pipeline with per-transmission tracking and improved chunk handling.
• Backward-compatible network receive path for older packet format support.
• Improved recording flow with pre-speech capture and smarter silence handling.
• Persistent audio cache with per-player storage and players index.
• Safer runtime behavior with additional validation, stale transmission cleanup, and better diagnostics.
• Expanded configuration options for playback radius, persistence, filters, and debug logging.
• Host-authority sound sync — greedy set-cover algorithm selects the minimum number of enemies needed to cover all nearby players. Host decides the voice filter mode (none / pitch down ×0.5 / pitch up ×1.2 / alien) and broadcasts a SyncPlayCommandPacket so every client plays the exact same clip with the exact same effect simultaneously.
• Custom audio folder — add your own .mp3 or .wav files without writing any code; mod authors can also register clips at runtime via MimicsAPI.
• Smart director and event reactions — context-aware selection of voice, enemy and listeners, plus reactions to deaths, damage, spotted players, valuables and the Head Grabber.

🛠️ Build from Source

1. Optionally point the build at your game install in TFS_Mimics.local.props (without it the build uses the R.E.P.O.GameLibs.Steam NuGet package, which can lag behind the current game version):

<Project>
  <PropertyGroup>
    <GameDir>C:\Program Files (x86)\Steam\steamapps\common\REPO</GameDir>
  </PropertyGroup>
</Project>

1. Build:

dotnet build TFS_Mimics.csproj -c Release

1. Output DLL:

bin/Release/TFS_Mimics.dll

For one-command build and deploy, see README.build.md.

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

🙏 Credits

• Harmony - Harmony Patching Library
• BepInEx - BepInEx Mod Loader
• r2modman - r2modmanPlus Mod Manager
• Thunderstore - Thunderstore Mod Hosting
• Mimic - Original mod for R.E.P.O v0.3.0 and lower by eth9n
• R.E.P.O - Game by semiwork
• ToxesFoxes - Mod development
• Contributors - See GitHub contributors page for details

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🤖 AI Content Disclaimer

This project used AI-assisted tooling during development for debugging, bug fixing, and technical iteration support.

• AI assistance was used as a development aid, not as an autonomous publisher.
• Final implementation decisions, integration, and testing were performed manually by the author.
• This notice is provided for transparency regarding the development workflow.