The scanner is designed to help answer
- What language or stack does this look like?
- How much code is being inspected?
- Are there obvious documentation gaps?
- Are there possible hardcoded configuration risks?
- Are there complexity signals?
- Are there migration blockers worth reviewing?
- What should be examined before refactoring or recoding?
Basic workflow
Create Project → Choose Scan Goal → Paste Code or Upload Allowed File → Run Private Preview Scan → Review Findings → Generate Report → Download Report → Submit Feedback
What the scanner may inspect
- estimated lines of code
- blank lines and comment density
- TODO / FIXME / HACK markers
- long lines and deeply nested blocks
- large files
- hardcoded URLs and IP addresses
- possible credential-like patterns
- database connection string patterns
- SQL string concatenation
- eval / exec-style patterns
- missing documentation signals
- deprecated syntax indicators where simple
What the scanner does not do
- execute code
- compile code
- run shell commands
- install dependencies
- guarantee vulnerability detection
- guarantee modernization success
- store raw source code by default
- display raw secrets
- modify production files
Decision categories in reports
Reports may eventually classify next steps as:
- Rebuild
- Recode
- Contain
- Reject
- Needs deeper review
“Contain” is a roadmap decision category and is not currently implemented in reports. It means a risky legacy system may first be isolated, wrapped, sandboxed, or profiled while a safer modernization path is planned — rather than being rewritten immediately.
Secret handling
Possible secrets should be redacted. A result may say "Possible secret-like value detected near line X," but it should not display the secret value.