Vaner

Status: alpha (pre-1.0). Interfaces may evolve quickly while we stabilize core behavior.

Vaner is a local-first predictive context engine for AI coding workflows. It uses idle time to anticipate likely next prompts, pre-build useful context, and serve the best context package quickly when the real prompt arrives.

Demo

CLI walkthrough:

vaner init --path .
vaner daemon start --no-once --path .
vaner query "where is auth enforced?" --explain --path .
vaner inspect --last --path .
vaner why --list --path .

Asciinema demo: coming soon.

Install

From source (current):

git clone https://github.com/Borgels/vaner.git
cd vaner
pip install .

One-liner (coming soon - requires v0.1.x on PyPI):

curl -fsSL --proto '=https' --tlsv1.2 https://vaner.ai/install.sh | bash

Installer source for review: scripts/install.sh.

Quickstart

vaner init --path .
vaner daemon start --no-once --path .
vaner query "where is auth enforced?" --explain --path .
vaner inspect --last --path .
vaner why --list --path .

Documentation

Most documentation lives at docs.vaner.ai:

• Getting started: docs.vaner.ai/getting-started
• Integrations: docs.vaner.ai/integrations
• Configuration: docs.vaner.ai/configuration
• Architecture: docs.vaner.ai/architecture
• Security: docs.vaner.ai/security
• CLI reference: docs.vaner.ai/cli
• Examples: docs.vaner.ai/examples

Community

• Contributing guide: CONTRIBUTING.md
• Security policy: SECURITY.md
• Code of conduct: CODE_OF_CONDUCT.md
• Support channels: SUPPORT.md
• Examples: examples/

OpenSSF Best Practices

Vaner is preparing an OpenSSF Best Practices submission.

License

Apache-2.0. Copyright 2026 Borgels Olsen Holding ApS (VAT DK39700425).