Back to directory

Blades

A multimodal AI agent framework for Go, supporting custom models, tools, memory and middleware for multi-turn conversation, chain-of-thought reasoning and structured output.

Blades

Blades is a multimodal AI agent framework for Go, built for multi-turn conversations, chain-of-thought reasoning and structured output, among other use cases, with custom models, tools, memory and middleware. It is built around a small set of core components: an Agent that invokes models and tools, a Prompt for templated text with variable substitution, a Chain that connects agents or other chains into workflows, a pluggable ModelProvider, external Tools such as APIs, databases or the file system, Memory for continuous conversation context, and Middleware for cross-cutting concerns.

Its middleware design draws on the Kratos framework's middleware philosophy, letting capabilities such as observability and guardrails be added to an agent without changing the agent's own code, and its interfaces are meant to make swapping between different LLM providers straightforward.

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.