jamjet-labs/jamjet

v0.5.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 1mo AI Agents & Assistants

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

a2a agent-memory agent-safety agentic-ai ai-agents ai-governance

+13 more

ai-safety approval-workflows audit-log human-in-the-loop java llm llmops mcp model-context-protocol production-ai python rust workflow

Summary

AI summary

Engram adds a PostgreSQL backend, message store with four new MCP tools, and multi-provider LLM support.

Full changelog

What's New

Engram — Durable Agent Memory (major update)

Engram is JamJet's built-in memory layer for AI agents — temporal knowledge graph, hybrid retrieval, fact extraction, conflict detection, and consolidation.

PostgreSQL backend — production-ready alternative to SQLite, auto-creates tables, selected via --db postgres://...
Message store — save, list, and retrieve full conversation histories (4 new MCP tools: messages_save, messages_get, messages_list, messages_delete)
11 MCP tools total (up from 7)
Improved fact extraction — rewritten extraction prompt for better recall of specific details (names, numbers, dates, relationships), increased candidate retrieval (50 to 100), longer timeouts for large payloads
Multi-provider LLM backends — Ollama, OpenAI-compatible, Anthropic, Google Gemini, shell-out command — one binary, no recompile
REST API — 13 endpoints via --mode rest
Python + Java SDK clients for Engram
Docker image on GHCR: ghcr.io/jamjet-labs/engram-server:0.5.0
Published on crates.io, GHCR, Maven Central, and the Official MCP Registry

Spring Boot Starter

engram-spring-boot-starter 0.2.0 — drop-in Spring AI ChatMemoryRepository backed by Engram

Java SDK

Bumped to 0.5.0 with EngramClient (8 blocking methods for memory + message operations)

Python SDK

Bumped to 0.5.0 with async EngramClient
--output json flag for jamjet run CLI command

Examples

claims-processing — insurance claims pipeline with 4 specialist agents (intake, assessor, fraud analyst, resolution writer), HITL approval, audit trails, durable execution

Other

README rewritten — updated Engram to 0.5.0, trimmed from 538 to 260 lines, added full examples table
AI-assisted contributions policy added
Docker fix for Glama sandbox compatibility

Install

pip install jamjet==0.5.0

cargo install jamjet-engram-server

docker pull ghcr.io/jamjet-labs/engram-server:0.5.0

Full Changelog: https://github.com/jamjet-labs/jamjet/compare/v0.4.0...v0.5.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track jamjet-labs/jamjet

Get notified when new releases ship.

About jamjet-labs/jamjet

Durable, agent-native AI runtime with native MCP client + server and A2A support. Rust core for performance, Python authoring for ergonomics. Features graph-based workflows, durable execution, and multi-agent coordination.