# JarvisCore Open-source Python framework for autonomous multi-agent systems with durable memory and production observability. - Findings24: - Site: - Preços: Freemium - Categorias: Infraestrutura para agentes de IA ## O que é JarvisCore? JarvisCore is an open-source Python framework for building autonomous multi-agent AI systems. It provides two execution profiles: AutoAgent, which runs an internal observe-orient-decide-act loop, and CustomAgent, which exposes the execution loop for more deterministic control. Both profiles use the same underlying runtime infrastructure. The framework is designed for agents that need to operate across extended workflows rather than produce a single response. Its memory model combines a working scratchpad, episodic ledger, compressed long-term summaries, and optional cross-session semantic memory through Athena MemOS. Agents can coordinate through a peer-to-peer mesh using SWIM gossip and ZMQ, allowing the same code to run in one process or across distributed machines. JarvisCore also connects agents to external services through typed Python integrations called atoms. The documented catalog includes services such as Slack, GitHub, Zoom, SAP, NetSuite, Microsoft Graph, and Salesforce, while the Nexus credential layer manages OAuth2, API keys, and basic authentication outside the agent's reasoning context. Tool calls, agent turns, and language-model requests are traced automatically, with Redis PubSub, JSONL, and Prometheus outputs described for streaming, compliance, and operational dashboards. The project includes workflow orchestration, crash recovery, and human review controls. In the financial intelligence example, three AutoAgent specialists fetch market data, analyze it, and produce a Markdown briefing in dependency order; rerunning the workflow skips completed steps. A quickstart requires installation of the Python package, initialization, an LLM provider key, and a dependency check before running an agent. ## O que JarvisCore pode fazer? ### Two agent execution profiles AutoAgent runs a full cognitive loop internally, while CustomAgent exposes the loop for deterministic control. Both profiles share JarvisCore infrastructure. ### Four-tier memory Agents can use a working scratchpad, episodic ledger, compressed long-term summaries, and optional cross-session semantic memory through Athena MemOS. ### Peer-to-peer agent mesh Agents discover and message one another through a SWIM gossip mesh and ZMQ. Nodes can join, fail, and rejoin without a central orchestrator. ### Typed service integrations Integrations are provided as typed Python functions called atoms. The documented catalog includes Slack, GitHub, Zoom, SAP, NetSuite, Microsoft Graph, and Salesforce. ### Credential isolation and tracing Nexus manages OAuth2, API keys, and basic authentication outside agent reasoning, while agent turns, tool calls, and LLM requests are traced through supported observability outputs. ## Casos de uso ### Sequential financial analysis Build a dependency-ordered pipeline in which separate agents fetch market data, analyze signals and alerts, and generate a Markdown briefing. The documented example also demonstrates resuming after a crash. ### Distributed research teams Use AutoAgent or CustomAgent profiles for research and analysis tasks that require multiple agents to communicate across a mesh rather than relying on one central coordinator. ### Operations across connected services Define an operations agent with capabilities such as team messaging, issue management, and documentation, then let the runtime search its typed integration registry for suitable tools. ### Long-running autonomous work Use durable memory, explicit failure reporting, observability, and human-in-the-loop review for workflows that need to retain context and support intervention over time. ## Perguntas frequentes ### What is JarvisCore? JarvisCore is an open-source Python framework for building autonomous multi-agent AI systems. It includes agent execution profiles, memory, peer-to-peer communication, integrations, credentials, workflows, and observability. ### What is required to run the quickstart? The documented quickstart installs the jarviscore-framework package, runs jarviscore init, adds an LLM provider key to the environment, and uses jarviscore check to verify dependencies before running an agent. ### How do AutoAgent and CustomAgent differ? AutoAgent runs an internal observe-orient-decide-act loop. CustomAgent exposes the execution loop directly for deterministic control. Both use the same shared infrastructure. ### Can workflows resume after a failure? The financial pipeline example uses a deterministic workflow ID and stores completed step outputs. When the example is rerun, completed steps are skipped and execution continues from the unfinished step. ## Informações rápidas - **Category**: Open-source AI agent framework - **Primary language**: Python - **Agent profiles**: AutoAgent and CustomAgent - **Communication model**: Peer-to-peer mesh using SWIM gossip and ZMQ - **Documented integrations**: Slack, GitHub, Zoom, SAP, NetSuite, Microsoft Graph, Salesforce, and more - **Example output**: Markdown financial briefing produced by a dependency-ordered workflow