Godot Release Dashboard Kit

godot-release-dashboard-kit builds a small static dashboard from JSON, Markdown, and image artifacts produced by Godot Production Toolkit commands.

It is intentionally simple: point it at a reports folder and it writes a self-contained HTML file suitable for a CI artifact or release checklist. Report cards are grouped into practical workflows and keep their release readiness state visible, so export, runtime evidence, mobile UI, content, and other checks stay easier to scan in one page. Common toolkit reports also show typed highlights such as frame p95, pack order, export preset counts, and risk scores without making readers open every JSON file first.

Install

python -m pip install godot-release-dashboard-kit

From a source checkout:

python -m pip install -e .\godot-release-dashboard-kit

Quick Start

godot-release-dashboard build reports\godot-project-doctor --output reports\dashboard.html

Write a machine-readable dashboard summary:

godot-release-dashboard build reports\godot-project-doctor --format json --output reports\dashboard.json

Compare the current reports with a previous report folder:

godot-release-dashboard build reports\current --previous-reports-dir reports\previous --title "Release Candidate Evidence" --description "Android export and runtime checks" --project "Demo Game" --output reports\dashboard.html

Include scenario run evidence in the same dashboard:

godot-scenario-report bundle reports\scenarios --manifest scenario-manifest.json --telemetry reports\runtime-timeline.json --evidence log=reports\run.log --evidence junit=reports\junit.xml --format json --output reports\release-evidence\scenario-bundle.json
godot-release-dashboard build reports\release-evidence --output reports\dashboard.html

Inputs

The dashboard scans a folder recursively for .json, .md, .png, .jpg, .jpeg, .svg, and .webp files. Toolkit JSON reports are summarized through their tool, kind, and summary fields when available. If a JSON report includes command, commands, tool_version, schema_version, generated_at, profile, risk, workflow, or category, the dashboard shows those fields as report metadata, reproduction commands, and grouping labels. workflow and category can appear at the report top level, inside metadata, or inside summary. When they are absent, the dashboard uses a conservative fallback based on the tool name, report kind, and file path. Image artifacts such as mobile UI overlays, screenshot diffs, pixel previews, and visual smoke captures are embedded into the self-contained HTML output. For common toolkit report shapes, dashboard cards include a small Highlights section with typed values such as runtime sample counts, frame p95/max, pack counts, pack load order, export preset counts, asset counts, and risk levels.

Scenario bundle JSON from godot-scenario-report-kit is shown as a release evidence card with scenario pass/fail counts plus the nearby files a reviewer should open next, such as logs, JUnit XML, runtime telemetry, profiler captures, visual-smoke reports, or screenshots. The dashboard shows link metadata from the bundle; it does not run the game or rewrite those evidence files. If the bundle contains a compact telemetry summary, the dashboard also shows sample count, frame p95, frame max, memory max, and budget spike counts.

When --previous-reports-dir is supplied, the dashboard scans the previous folder with the same report rules and adds a compact change section. The trend cards show added, removed, and changed report cards, including status changes and error/warning deltas. The --baseline flag is accepted as a shorter alias.

Outputs

• html: self-contained static dashboard with release readiness metrics, source report links, workflow-grouped report cards, scenario evidence sections, and image previews.
• json: summary for scripts or later dashboard tooling, including counts for blocked, attention, ready, workflow groups, scenario bundles, scenarios, and linked scenario evidence. Reports that include reproduction commands are counted in summary.reports_with_commands, and report cards can include highlights rows for typed summary values. Scenario-linked telemetry summaries are rolled up as sample, spike, warning, and error counts. When a previous folder is supplied, JSON output also includes previous_summary, trends, and trend-related summary counts.