Urgent.News

What's breaking now, across thousands of outlets.

AI

OpenAmer ASI Core: 5 native tools, a 10-subsystem heartbeat, and an A2A Global Mesh

Most agent frameworks glue everything together with subprocess calls and a pile of schedulers. We took a different path with OpenAmer (Apache 2.0): an in-process ASI core, one heartbeat instead of 84 cron jobs, and an agent-to-agent mesh where every instance talks to every other instance. 1. Five native ASI tools - no subprocess The ASI core ships five tools that run in-process . No…

OpenAmer ASI Core is a novel agent framework that aims to simplify and streamline the process of managing multiple agents. Unlike traditional frameworks that rely on subprocess calls and numerous schedulers, OpenAmer adopts an in-process ASI core, a single heartbeat, and an agent-to-agent mesh communication approach.

The ASI core houses five native tools that run directly within the process, eliminating the need for subprocess.run, JSON-RPC communication, and reducing startup latency per call. The CLI interface allows users to check the status of the ASI core, perform reasoning, learn from experiences, remember past interactions, trigger subsystem events, and view the overall heartbeat status.

One of the key innovations in OpenAmer is the 10-subsystem heartbeat. Instead of relying on 84 separate cron jobs that handled various tasks such as browser watchdogs, outreach, self-healing, learning loops, wiki generation, and backups, OpenAmer consolidates these functions into a single heartbeat. This heartbeat triggers ten subsystems, each providing its own health signal.

As a result, there is only one place to monitor the overall health of the system and one place to perform repairs, eliminating the need to identify which of the 84 cron jobs caused any issues.

Another noteworthy feature of OpenAmer is the A2A Global Mesh. In this communication model, every instance of OpenAmer communicates directly with every other instance, rather than relying on a central broker. When two instances are run on a local area network (LAN), they can coordinate and exchange agent-to-agent messages directly. This mesh topology ensures that the system remains functional even if a single node fails, as the other instances can still communicate and operate independently.

OpenAmer has been designed with extensive testing and validation. All 16 ASI capabilities covered by the framework have been verified through executable checks, and the entire system runs locally on Windows, making it suitable for laptop-first agents. The framework is released under the Apache 2.0 license and is available on GitHub at https://github.com/openamer/openamer.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Built for the agent at the keyboard

A coding agent cannot read a progress bar. It reads stdout, parses it and decides what to do next. If the answer is buried in a frame meant for human eyes, the agent guesses.

  • vx coding agent parses stdout to determine next actions
  • JSON output includes run ID, status, and task details
  • vx maintains history of 50 recent failed runs in SQLite file

More from Friday 9 October →