# Example Artifact Pack
When you run `prview`, it generates an artifact pack in `~/.prview/runs/<repo>/<branch>/<run_id>/` (or `$PRVIEW_HOME/...`).
New run ids use a timestamp plus short HEAD suffix, for example `20260704-120500-a1b2c3d`; treat the value as opaque.
This pack contains all the analysis output in both human-readable and machine-readable formats.
## Directory Structure
A typical artifact pack looks like this:
```
├── PR_REVIEW.md # Unified review narrative with risks and recommendations
├── report.json # Full machine-readable report payload on disk
├── 00_summary/
│ ├── RUN.json # Run metadata + execution mode + check inventory
│ ├── FAILURES_SUMMARY.md # Compact blocking failures with links to logs
│ ├── MANIFEST.json # SHA256 hashes for generated files
│ ├── SANITY.json # Integrity validation results
│ ├── MERGE_GATE.json # Machine-readable merge decision
│ ├── MERGE_GATE.md # Human-readable merge decision
│ ├── pr-metadata.txt # Branch/base/profile metadata
│ ├── file-status.txt # A/M/D + file paths
│ └── commit-list.txt # hash date author message
├── 10_diff/
│ ├── full.patch # Full diff with diff-stat header
│ ├── per-commit-diffs/ # Individual commit patches
│ └── per-file-diffs/ # Hotspot files (>=80 lines changed)
├── 20_quality/
│ ├── *.result.json # Per-check machine-readable outputs
│ ├── *.log # Per-check raw logs
│ ├── coverage-delta.txt # Source↔test mapping with change status
│ └── BREAKING_CHANGES.md # Removed pub symbols, changed signatures
├── 30_context/
│ ├── INLINE_FINDINGS.sarif # Optional SARIF output for findings
│ ├── cargo-tree.txt # Dependency tree
│ ├── cargo-sbom.json # Generated SBOM
│ └── npm-sbom.json # Generated SBOM
├── dashboard.html # Visual HTML summary
└── artifacts.zip # Everything zipped
```
## Key Files
1. **`PR_REVIEW.md`**: The main entry point for a human reviewer. It contains a narrative summary of the PR, including structural risks, test coverage gaps, and architectural insights.
2. **`00_summary/MERGE_GATE.json`**: The canonical source of truth for CI/CD automation. It determines if the PR is safe to merge based on the active `.prview-policy.yml`.
3. **`dashboard.html`**: A zero-dependency, self-contained HTML dashboard that visualizes the PR metrics, test hotspots, and quality gates.
4. **`report.json`**: The complete state of the analysis, useful for building custom integrations or training AI models.