Back to directory

Agno-Go

A multi-agent framework written in Go that follows the design of the Python Agno project, providing agents, teams and workflows with multi-provider model support and an HTTP server for running them as a service.

Agno-Go

Agno-Go builds agents, teams with four coordination modes, or workflows out of five primitives, and lets these be mixed freely; teams inherit or default their models, and workflows can resume from saved snapshots. Its AgentOS component is a REST server with an OpenAPI 3.0 specification, session storage, health checks, structured logging, CORS support, request timeouts, and parity endpoints for run summaries, reuse and history filtering, matching the behavior of the original Python runtime including shared sessions, asynchronous and synchronous summaries, and streaming-event flags.

It connects to many model providers, including OpenAI, Anthropic Claude, Google Gemini, DeepSeek, Ollama, Cohere, Groq, Together and OpenRouter, and includes tools for calculations, HTTP requests, file operations, search, Claude Agent Skills, and integrations such as Jira, Gmail and ElevenLabs speech, plus ChromaDB-based knowledge and retrieval-augmented generation. Guardrails include a prompt-injection guard, custom hooks, and media validation, and it emits a server-sent-event stream with reasoning snapshots for observability.

It is written in Go and released under the Apache-2.0 license.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.