Ross ROSS = Recommend OSS · open-source software intelligence for agents

Strands Agents

Build an agent harness and control it end-to-end. Open-source SDK for production AI agents in Python & TypeScript - any model, any cloud. observed · 2026-08-28

github.com/strands-agents/harness-sdk · homepage · Python · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

81/100

  • Activity 99
  • Release rhythm 86
  • Longevity 34
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 6
  • age_days: 476
  • days_rel: 12
  • days_push: 7
  • n_releases_24m: 88

Full methodology

Adoption not part of the score

7022 stars · 1068 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Strands Agents is an open-source SDK (Python and TypeScript) for building production AI agents with a model-driven approach, supporting any model provider including Amazon Bedrock, OpenAI, Anthropic, and Ollama. It provides an agent loop, custom and built-in tools, MCP support, lifecycle hooks, multi-agent patterns (swarms, graphs, workflows), session management, and OpenTelemetry observability.

Use cases

  • build an AI agent in Python with a few lines of code
  • create autonomous agents that call custom tools
  • connect agents to MCP servers and tools
  • orchestrate multi-agent systems with swarms and graphs
  • swap between Bedrock, OpenAI, Anthropic, and Ollama model providers
  • deploy agents to AWS Lambda, Docker, or AgentCore
  • add structured output and validation to agent responses
  • monitor and trace agent behavior with OpenTelemetry

When to choose

  • you want a production-ready agent SDK with both Python and TypeScript support
  • you need model-provider flexibility across clouds
  • you want built-in multi-agent orchestration, hooks, and session persistence
  • you plan to deploy agents on AWS infrastructure
  • you need MCP integration and a large ecosystem of community tools

When to avoid

  • you need a no-code or GUI agent builder rather than a code SDK
  • your project is limited to a single non-supported model provider
  • you want a lightweight prompt library rather than a full agent harness
  • you require stable long-term APIs and cannot tolerate frequent releases and evolving features

Facets

library · maturity active

agent-framework llm-inference mcp sdk workflow-automation chatbot large-language-models artificial-intelligence developer-tools python cross-platform serverless cloud multi-agent-systems model-providers bedrock openai anthropic agent-loop hooks swarms graphs structured-output observability opentelemetry typescript-sdk python-sdk production-agents ai-agents automation nodejs docker

10 sources

Member repositories

RepositoryRoleHealth v2
strands-agents/harness-sdkmain81
strands-agents/toolsplugin86

For agents

markdown · JSON · MCP: product_card(name="strands-agents/harness-sdk")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem