FREE-FIRST. NO HIDDEN INFERENCE.
Start with evidence.
Bring the intelligence you need.
The application has no billing system. You control provider usage and infrastructure. Free service tiers are subject to their current limits and eligibility.
Static analysis
Evaluation no billing integration
Run Intervexa on your own infrastructure.
- GitHub App repository ingestion
- Bounded Python symbol analysis
- Revision-pinned findings
- Advisory GitHub Checks
OPTIONAL INTELLIGENCE
Bring your own key
Usage billed by your provider
Choose OpenAI or OpenRouter. No shared paid key and no automatic switching.
- Ephemeral credentials
- Verified model metadata
- Explicit generation consent
- Bounded output token budget
Isolated execution
Your host
Operate a separate Linux runner on infrastructure you already own or authorize.
- Rootless disposable containers
- Network-disabled test execution
- Baseline and PR comparison
- Exact merges and retained logs
AI inference and hosted infrastructure may incur charges from third parties. No unlimited free inference is included. Connection tests do not generate content; generation requires explicit consent.