RunAudit – Playtest Analytics and Game Balance Toolkit

Turn playtests into evidence.

RunAudit records the gameplay data you choose and brings completed runs together in the Unity Editor. See which upgrades players pick, where runs end, and how outcomes differ between builds—without setting up a backend, database, account or telemetry service.

Built for developers iterating on run-based games, roguelites, survival games and repeatable playtests. You define the metrics, metadata, events and outcomes that matter to your game.

Follow a playtest as it happens

The Live Dashboard shows run status, active and paused time, numeric metrics, metadata and recent events. Recording supports explicit pause/resume, timeScale pauses, scene changes and safe finalization on normal application quit.

Understand completed runs

Filter local history, classify results, and inspect success rates, durations, numeric metrics, metadata groups, ending reasons and recorded build versions. Events/items analysis shows pick rates and success with and without an event detail. Counts and denominators stay visible so you can judge the evidence. These are associations, not proof that an item causes an outcome or automatic game balancing.

Bring external playtests into the same workflow

Preview supported RunAudit reports before importing. Label batches and tester/source provenance, inspect duplicates and conflicts, then analyze local and imported runs together or separately. Reports move through files you choose; RunAudit does not send or retrieve tester data over a network.

Integrate through UnityEvents or C#

Five no-code action components handle lifecycle, metrics, events, metadata and 3D trigger relays. Wire them to existing gameplay UnityEvents, or use the direct C# API with using RunAuditTool; and calls such as RunAudit.StartRun(), Add(), Set(), Meta(), Event() and EndRun(). No-code components require explicit integration; RunAudit does not discover game variables automatically.

Keep usable files

Completed sessions produce Markdown and CSV reports you can inspect, archive or share yourself. The dashboard opens or reveals files using your operating system’s applications. RunAudit recording and analysis stay local and work offline.

Optional playable example

Import Beacon Survival from the package’s Samples tab to explore a working integration. Its explicitly generated demonstration data stays in a separate sample library. The core does not depend on the sample. Local documentation covers setup, the API, integration recipes, report storage and troubleshooting.

Compatibility and boundaries

Validated with Unity 6000.4.7f1 on Windows, including Windows x64 player recording. Dashboard and historical analysis are Editor tools. Other Editor versions and operating systems are not claimed as tested. WebGL builds were compiled, but browser execution, report persistence and browser shutdown are unvalidated. Core has no render-pipeline dependency; Beacon uses Built-in materials and legacy input and may need setup in other templates.

No third-party package or paid-service dependencies. Standard Unity modules are required. RunAudit supports one main-thread session at a time, records explicitly integrated actions, and does not preserve an unfinished session through assembly recompilation, hard crashes or forced termination. File storage depends on platform permissions. Intended for local development playtests, not cloud live-ops analytics.

Support

Product and support website: https://seighler.pages.dev/#support

Contact mashsaylor@hotmail.com for installation, documented integration or reproducible defects. Include RunAudit and Unity versions, operating system, reproduction steps, Console messages and a minimal example when possible. Remove private tester data before sharing reports. No fixed response-time commitment is offered.

AI assistance disclosure

OpenAI Codex assisted development of RunAudit’s C# runtime and Editor code, no-code components, optional sample, procedural UI styling, automated tests and documentation. The developer directed scope, workflows and visual requirements through iterative review. Generated implementation was modified through product-specific integration, lifecycle hardening, report validation, assembly and namespace organization, and repeated automated and actual Unity Editor testing. Release preparation includes fresh-install tests, player builds and genuine UI captures. Listing text and screenshot compositions were also prepared with Codex assistance. Marketing screenshots use actual captures with external crops, captions and layout, not generated replacement UI. RunAudit itself provides no AI analysis and calls no AI service at runtime.