See CodeDelta run
Two versions of a codebase in: every change, every AI agent and every attack out. Real runs on Kubernetes, the Godot engine and Apache Commons Lang.
11 sections · 67 screens · under 6 minutes in all, 15 to 50 seconds a section · or run any section below
| What it does | ||
|---|---|---|
| Churn measurement | Exactly what changed between two snapshots of the code: changed, deleted and added logical statements per file and per language, at any scale. | |
| TRUE_CHURN & data metrics | Generated files subtotalled out of the headline number; working code split from data; REWORK and REP_CHURN, the share of a change that was wholesale replacement. | |
| Code Browser & diff view | The side-by-side diff with every statement classified and coloured, the files, the classes, and the include map drawn as a picture. | |
| AI Agent Scan | Every place your code calls AI: agent SDKs, model endpoints, exec-on-model-output, committed credentials, agent infrastructure; file and line for each, drawn as the Agent Map. | |
| Threat Index | What in a change looks like an attack: the same new code added to many files, obfuscated code, a large new file in an otherwise quiet release. | |
| Build-File Alert | Every build, CI, packaging and dependency file that changed between two versions, with install hooks flagged at the top. | |
| Trend database & summary | Every run accumulates in a local database; a one-page summary charts churn across releases. | |
| AI Agent Tracer | What your coding agents did, read from the session logs they leave on the machine: what each ran, deleted, pushed and sent, under which model. | |
| AI Log Keeper & token consumption | Free download: saves the agents’ session logs before they are deleted, and counts token consumption per user, team and project against budgets you set. | |
| GitHub, GitLab & MCP | The same engine on every pull request, in GitLab merge requests, from the command line, and inside AI assistants through the MCP server. |