Only in NirmIQ

Your code knows more than your documents say

Code Intelligence bridges the gap between what your developers build and what your safety engineers document. Scan repositories, generate requirements, create FMEA items, and detect drift — automatically.

// The Code Intelligence Pipeline
STEP 1
Connect your repository
GitHub, GitLab, or any Git remote
STEP 2
AI analyzes your code
Understands what your code does and where the risks are
STEP 3
Requirements generated
Structured requirements with hierarchy and traceability
STEP 4
FMEA failure modes created
Risk analysis items linked to code and requirements

The gap between code and documentation is where failures hide

Developers write code. Safety engineers write requirements. They rarely look at each other’s work. When they drift apart, that’s when things break in production.

Undocumented code

Features exist in code but have no matching requirement. If it’s not documented, it doesn’t exist for compliance.

Stale requirements

Requirements say one thing, code does another. The drift accumulates silently until an audit or an incident.

Missing risk analysis

New code paths introduce new failure modes. Without code-to-FMEA linking, risks are discovered in testing — or production.

Scan any codebase

Point Code Intelligence at your repository. Within minutes, you’ll know what your code actually does — not what the docs say it does.

  • Multi-language support

    Python, C/C++, Java, JavaScript, TypeScript, Go, Rust, and more

  • Deep understanding

    Goes beyond surface-level scanning. Understands what your code does, not just what it looks like.

  • Quick scan mode

    Get results in minutes, not hours. Scan the whole repo or focus on specific directories.

# Code Intelligence scan results
✓ Scanned 847 files across 12 modules
✓ Identified 234 functions with safety implications
⚠ 47 functions have no matching requirement
✗ 12 requirements reference deleted code paths
→ Generated 47 new requirement suggestions
→ Generated 31 FMEA failure mode suggestions
GENERATED FROM CODE
REQ-AUTO-001
System shall validate all sensor readings against calibrated thresholds before processing
Source: sensor_manager.py:validate_reading()
REQ-AUTO-002
System shall execute emergency shutdown sequence within 100ms when critical temperature threshold is exceeded
Source: safety_controller.py:emergency_shutdown()
REQ-AUTO-003
System shall log all state transitions with timestamp and triggering event for audit trail
Source: state_machine.py:transition()

Generate requirements from code

Your code already encodes what the system does. Code Intelligence turns that into structured requirements — with hierarchy, types, and traceability back to the source.

  • Structured output

    Generated requirements include type classification, hierarchy level, and suggested custom IDs

  • Source traceability

    Every generated requirement links back to the exact file, class, and function it came from

  • Review before commit

    AI suggests, you review and approve. Nothing gets added without your sign-off.

Generate FMEA from code

Code paths reveal failure modes that documents miss. A function that handles sensor data can fail in ways that only become obvious when you look at the code.

  • Code-derived failure modes

    Discover failure modes that only become visible when you look at the actual implementation

  • Pre-filled S/O/D ratings

    AI suggests severity, occurrence, and detection ratings based on code context

  • Linked to requirements

    Generated FMEA items automatically link to both the source code and related requirements

FMEA GENERATED FROM CODE
RPN: 216High Priority
Sensor data timeout not handled in shutdown path
S:9 O:4 D:6 | Source: safety_controller.py:88
RPN: 120Medium Priority
Integer overflow in temperature conversion
S:8 O:3 D:5 | Source: sensor_manager.py:45
RPN: 60Low Priority
Race condition in concurrent state transitions
S:6 O:2 D:5 | Source: state_machine.py:112

Detect code drift before it becomes a problem

Requirements say “shall respond within 100ms.” The code now takes 250ms after last sprint’s refactor. Code Intelligence catches this before your auditor does.

Deleted code paths

Requirements reference functions that no longer exist. Instant visibility into broken traceability.

Changed behavior

Code logic changes that affect the intent of a requirement. Flagged automatically on every scan.

New untraced code

New functions and modules added without corresponding requirements. Caught before the next audit.

Built for teams that ship code and prove compliance

Safety Engineers

Get visibility into what developers built, automatically linked to your FMEA and requirements.

Developers

Stop manually writing requirements for code you already wrote. AI generates them from your implementation.

QA & Test Teams

Know exactly which code paths need test coverage based on requirement and FMEA links.

Compliance Teams

Prove bidirectional traceability from code to requirements to risk analysis in one click.

Stop guessing what your code covers

Code Intelligence is available on the Power tier and above.

Start with a free trial. Connect your first repository in under 5 minutes.