Back to directory

lovia

A restrained Python framework for building AI agents, covering tools, sessions, events, context compaction, and serving with a small set of composable primitives rather than a large platform.

lovia

lovia is a Python framework for developers who want to own the agent loop without rebuilding tools, sessions, events, context compaction, and serving from scratch, while avoiding a large stack of abstractions. Its core depends only on httpx, pydantic, and pyyaml, and installs everything else on demand.

An Agent is immutable configuration, typically varied with clone(); a Runner executes one run; the @tool decorator wraps a typed callable; and handoff or agent-as-tool patterns compose multiple agents, with skills, MCP, a todo list, and memory implemented as plugins on the same extension point. OpenAI, Anthropic, and OpenAI-compatible endpoints are built in, and context compaction changes only the next model view without rewriting the stored conversation, keeping prompt prefixes stable for caching.

Installing the web extra serves a full chat UI with memory, skills, and scheduling in one command.

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.