Skip to main content

AI-code audit

By 2027, 40-60% of code submissions will be AI-generated. Regulated teams need an answer to one compliance question: which findings sit on AI-touched code? Vulkro is the only SAST tool that ships a defender-side AI-origin audit today.

Three CLI surfaces use the audit:

  • AI-origin fingerprinting (per-tool markers: Claude, Copilot, Cursor, Aider, ChatGPT, generic AI-generated)
  • --ai-code-segregation segregation report
  • --attest-reviewed reviewer attestation

Why this is on the SAST tool​

Most security tools run the same detectors on every line. That is right in technique but misses a posture point: AI-generated code fails in specific ways more often - boilerplate auth that does not enforce, copy-paste secrets, missing error handling, hallucinated dependency names. The scanner already knows about each finding, so it is the right place to flag which findings sit on AI-touched code.

Vulkro stays on the defender side. This module audits AI-generated code; it does not generate fixes via a vendor LLM. That fits the offline-first posture: your code never leaves the host.

How origin is detected​

The origin scan looks at each module for canonical per-tool markers:

ToolMarkers it looks for
ClaudeGenerated by Claude, Generated with Claude, Co-Authored-By: Claude, @claude-generated, @anthropic-generated
GitHub CopilotGitHub Copilot, Copilot-generated, @copilot
Cursorcursor.sh, @cursor-generated
Aideraider:, // aider
ChatGPTGenerated by ChatGPT, OpenAI GPT-...
GenericAI-generated, Generated by AI, # generated by ..., // generated by ...

The marker scan is strict precision today: no explicit marker means the file is treated as hand-written. Heuristic shape detection (long boilerplate docstrings, uniform variable naming) is planned for later. The strict scan is the safer first step, since falsely tagging a hand-written file as Copilot output would be worse than the current under-coverage.

Marker line numbers are kept, so Vulkro can compute a per-file marker ratio ("3 of 280 lines carry an AI marker").

--ai-code-segregation​

Adds a markdown report after the scan output:

$ vulkro scan . --ai-code-segregation
... (normal scan output) ...

# Vulkro AI-Code Segregation Report

- Total findings: 42
- Findings on AI-touched files: 11 (26.2%)

## Per-tool breakdown

- claude: 7 finding(s)
- copilot: 3 finding(s)
- generic-ai: 1 finding(s)

## AI-touched files (marker-line ratio)

- src/api/users.ts: 0.71%
- src/api/auth.ts: 1.43%
- src/services/notifier.ts: 0.36%

The report goes to stderr. The main output stream (table / JSON / SARIF) is unchanged, so machine-readable formats stay parse-clean.

--attest-reviewed​

Stamps every finding on an AI-touched file with a human-reviewed-ai-code evidence row. A downstream compliance emit can then show "every AI-touched finding has a human-review sign-off" without tagging each finding by hand.

$ vulkro scan . --attest-reviewed --reviewer jane@team
... (normal scan output) ...
attest-reviewed: stamped 11 finding(s) on AI-touched files (reviewer: jane@team)

The reviewer name defaults to $USER; pass --reviewer NAME to override. The attestation lives on evidence[].signal = "human-reviewed-ai-code" and evidence[].detail = "Reviewer attestation: AI-touched code reviewed by ...", so it carries into SARIF, JSON, JUnit, and the database emit.

Use cases​

HIPAA / PCI-DSS / FedRAMP audit evidence​

The auditor asks you to show that every AI-touched line was reviewed by a human. Run vulkro scan . --ai-code-segregation --attest-reviewed --format sarif > scan.sarif. The SARIF carries both the per-tool breakdown (properties rolls up segregation counts) and the per-finding attestation row. Hand the auditor the SARIF file.

Pre-merge gate on AI-shop PRs​

CI runs vulkro scan . --ai-code-segregation on every PR. The stderr report goes into the CI log. If the AI-touched share crosses the team's threshold, a wrapper script fails the build and posts the report as a PR comment.

Forensics after a security incident​

A security event surfaces in production. Run vulkro scan . --ai-code-segregation against the commit that shipped the bug. The output tells you which tool produced the offending code, so the team can update its reviewer playbook for that tool.

Last updated on