Running one
Iris Code: Scan Workspace from the command palette, or the Scan Workspace button in the Workspace tab. It walks every supported file and gives you:- Readiness - the top-level score, blockers, warnings, TODO load and trend
- The totals - files, directories, lines, functions
- A language breakdown
- Hotspots - the five biggest files and the five most complex
- Unused packages - declared in
package.json,go.mod,requirements.txtorpyproject.toml, never imported - A CVE badge on the Dependencies section, green when the last cached dependency scan found nothing, otherwise the count and worst severity. It reads the local cache only and never goes to the network on its own. See Dependents Table
- Issues - every blocker and warning in one list
- TODOs - grouped by type
- A file tree you can click through
- Trends - the score delta plus your three most improved and three most regressed files
The free review omits unused packages, dependency conclusions and trends. Those answers depend on a current scan, and reporting them from a snapshot taken weeks earlier would state something no longer true. Refresh With Pro replaces the rescan button, and the saved evidence stays viewable and clickable.
Just one folder
Right-click any folder in the Explorer and pick Iris Code: Scan This Folder. Or hit Scan Current Folder in the Folder tab to scan whatever the open file lives in. The Folder tab shows the same stats, chart, rankings and tree as the workspace, with its own Rescan and Clear.One full workspace scan is free without an account. Rescanning and folder scans are Pro. File reviews and click-to-open stay free forever.
Scanning again
A second scan of the same workspace or folder in VS Code or a JetBrains IDE does not analyse every file from scratch. Iris Code reads each file again and reuses its earlier analysis when the file’s contents, the settings that apply to it, its path and its baseline status are all unchanged. Everything else is worked out again from the current files: the totals, the workspace score, unused packages and duplicate findings. Duplicates always compare every current file, so code copied from an unchanged file into an edited one is still found. When a scan finishes, a notification says how many files were analysed and how many were reused, for example “Analysed 1 file, reused 3 unchanged.” A change that affects every file, such as a newduplicateBlockMinTokens, analyses everything again, which is expected.
Rescan everything in the Workspace tab, the folder view or the command palette ignores the reused results and analyses every file.
Reused results are kept in memory for the editor session, within a fixed memory budget, and are discarded when the editor closes. The CLI, git hooks, CI, the MCP server and cloud scans analyse every file on each run.
Issues and TODOs
Both tabs fill from your most recent scan, whether that was a folder or the whole workspace. A label at the top tells you which (“Showing: folder · src”), with one click back to workspace-wide, so you’re never reading folder results thinking they’re the whole project. Each row gives you a severity dot, the file and line, and what’s wrong. Click it to land there. The Issues badge shows errors in red, or the total when there are none. The TODOs badge counts everything. Findings stay put until you rescan or clear.Trends
After each workspace scan, Iris Code saves one score per file for that day. Over a few weeks that’s the difference between “I think it’s getting better” and knowing. In a file, you get a ↑ or ↓ next to the score against the last snapshot. In the workspace, the overall delta plus your three most improved and three most regressed files. How it works:- One snapshot per day. Scan five times today and you update today’s, rather than creating five
- They live in
.iris-snapshots/at your project root - That folder is added to
.gitignorefor you, because trends are yours and shouldn’t be committed - Delete the folder to start over
Exporting it
Turn the current analysis into a single dark-themed HTML file you can send someone, attach to a PR, or keep.1
Run a scan
File, folder or workspace. There has to be something to export.
2
Run the command
Iris Code: Export Scan Report as HTML from the command palette.
3
Pick where it goes
Defaults to
iris-report.html in your project root.4
Look at it
Open in Browser on the confirmation.
iris-report.html to the current directory. --output <path> puts it elsewhere, and you can pass a directory to scan.
Export is free on every plan. It’s a snapshot, so rerun the scan and export again for fresh numbers.