OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Evaluation & Observability

Meta Core

Meta Core is a Python telemetry layer for multi-agent AI systems. Its Transponder validates and appends structured events to a local JSONL stream, while a read-only observer reconstructs trace trees and flags handoff depth, cycles, orphan spawns and stale cores.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Meta Core — OpenViST project profile in Agent Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an agent system can emit structured events and you need local trace reconstruction and observer status signals without adding network or Git operations to the event-writing component.

Who it is for

  • Developers integrating telemetry into multi-agent systems
  • Operators who need to inspect agent traces and lifecycle events

Documented uses

  • Record core lifecycle, execution and spawn events in a shared stream
  • Reconstruct trace trees from agent events
  • Identify handoff depth, cycles, orphan spawns and stale cores
Before you choose

Requirements & limits

The repository is public. GitHub repository metadata identifies Apache-2.0, while the README's license section says “TBD”; check the repository's license files before relying on the metadata.

  • The Transponder writes locally and does not perform network or Git operations.
  • Git synchronization is handled separately by a batch utility; log rotation is also a separate shell utility.
  • The observer returns exit codes 0, 1 and 2 for OK, WARN and CRITICAL.
  • The broader Flow-Core, Control Plane and Chronos V3 design is described as roadmap rather than part of the working telemetry layer.
  • The README lists Python 3.10+, Bash, Git and util-linux flock as requirements.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Add a telemetry event and inspect the trace

Write an execution event to the local stream, then ask the observer to parse it.

Before you start

  • Python 3.10 or later
  • Bash, git and util-linux flock
  • A local Meta Core checkout

Steps

  1. Set CORE_ID and EVENT_LOG_PATH for the core and event stream.

  2. Run telemetry/transponder.py with a JSON event, or call the Python emit() API.

  3. Run telemetry/observer.py --json to read the stream and report its findings.

Expected outcome

The event is appended to the JSONL stream and the observer reports parsed trace findings in JSON.

Guide limitations

  • The observer analyzes the event stream; it does not itself provide a dashboard or route alerts.
  • Execution events require trace_id and span_id, which are generated as a root span when omitted.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
8 October 2026
Original website title
Sega757/meta-core · Official homepage title · retained in its source language
How we review sourcesSuggest a correction