SecureSense documentation
SecureSense is a VS Code extension that finds security vulnerabilities in Python code with two static analysers, checks every finding against your code, and tells you which ones are real — with the evidence.
What it does
Static analysis tools are noisy. A typical scan flags hundreds of lines, many of which are false alarms: the input was already sanitized, the code is never reached, or two tools reported the same bug twice. SecureSense sits between the scanners and you:
- Scans your workspace with Semgrep and CodeQL, side by side, on your machine.
- Correlates their output: duplicates from both tools become one finding, and findings that share one fix are grouped.
- Gathers evidence for each finding from your code's structure (its AST): does user input reach it, is it sanitized, who calls it, is it on a public web route, do its dependencies have known CVEs.
- Classifies each finding as Confirmed, Likely vulnerability, Inconclusive or Likely false positive, with a priority score and a rationale that cites the evidence.
- Lets you decide — approve, reject, or ask for investigation — in an in-editor dashboard, and optionally shares everything with your team on the web.
Where to go next
| If you want to… | Read |
|---|---|
| Try it in ten minutes | Getting started |
| Install the scanners and tools properly | Installation |
| Learn the panel, dashboard and decisions | The VS Code extension |
| Work with a team | Team sync & web dashboard |
| Understand how findings are merged | Scanning & correlation |
| Understand the verdicts and the AI's limits | Evidence & AI triage |
| Know what leaves your machine | Privacy & data |
| Fix a problem | Troubleshooting & FAQ |
SecureSense is a final-year project. It analyses Python (Flask, Django, FastAPI and plain scripts); other languages get the scanner results without the AST evidence.
Something unclear or wrong? Edit these docs on GitHub.