› cavi-ai/

Inspectable tools for local models, agent runtimes, and knowledge workflows.

Independent research and open source. Every release ships with its source, verification date, and commit, so you can check it before you adopt it.

Explore CAVIRead documentation
released artifacts
7
agent skills
31
licensed, source published
MIT

/explore

Everything CAVI ships

Seven released artifacts across editors, browsers, runtimes, and packages. Each one links straight to its documentation, source, and install path.

/solutions

Or start from an outcome

Each solution connects the job you want done to the products and skills that complete it.

01Apple Silicon · local inference

Run models locally

Find MLX models that fit your Apple Silicon host, verify them against an installed runtime, and wire a selected model into an agent workflow.

Open solution
03Browser runtime · audit trail

Browser automation with an audit trail

Give each agent a capability-scoped token, drive the browser through one runtime with shared idempotency and evidence contracts, and restart long workflows from a checkpoint instead of replaying side effects.

Open solution

/trust

Check the release before you adopt it

CAVI publishes source, verification dates, and immutable provenance for every featured release. Inspect the boundary, cross runtimes carefully, keep local control.

Featured releaseMLX Agent v0.5.2
License
MIT licensed
Verification
Verified Jul 2026