Jev in production › Agent context and memory

Astra (MatrixOrigin)

Self-hosted agent runtime with Jev as an optional judgment layer for memory filtering, lesson dismissal, request classification and skill selection, evaluated against a no-Jev arm on synthetic memory cases.

Open on GitHub ↗

The source publishes no measured number.
Use
Agent context and memory
Industry
AI infrastructure
Form
Open-source tool
Stage
Beta
Listed
2026-09-20
Found via
github
Repository
matrixorigin/Astra
Stars
34
Forks
11
Last push
2026-10-10
Language
Rust
License
Apache-2.0
Astra (MatrixOrigin) screenshot
docs/assets/explain-analyze-demo.gif in the matrixorigin/Astra README, Apache-2.0; shown from GitHub.

The README opens with

An agent runtime with native Jev judgments and EXPLAIN ANALYZE for context

Inspect context and state. Adjust and recover with evidence. Run in your environment.

Why Astra · Jev integration · Research · Quick start · Architecture · Runner · Comparison · Docs

Astra is a self-hosted runtime for long-running agent Work. Every model request is assembled by a budgeted pipeline you can EXPLAIN, every attempt leaves evidence you can inspect, diff, and roll back, and execution runs through a Runner inside your own environment.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for agent context and memory