Strands

In this catalog it is the Frameworks option when you need to compose multi-step agent loops and tool handoffs. Keep Strands on the observe/propose side of the loop: useful signals and drafts, with human or policy authority still required before anything irreversible runs.

Strands Agents is an open SDK for production agents with model-driven tool use and composable primitives.

1. Define the agent, model, and tool set.
2. Run; the model selects tools.
3. Tool results re-enter the model loop.
4. Emit the final response to the caller.

Related: agno · pydantic-ai · topics/12-central-nervous-system

• Use when: Strands Agents is an open SDK for building production-oriented agents with model-driven tool use and composable primitives matches the bottleneck and you specifically need to compose multi-step agent loops and tool handoffs.
• Skip when: another pack already covers the need with less ops surface, or Strands would become a second place people retype truth.
• Prefer instead: compare Agno, Pydantic AI before locking the cell.

• Ops: confirm license, hosting, and release cadence for Strands (strands-agents/harness-sdk) before relying on it in a live corridor.
• Integration: bad credentials, scopes, or schema drift produce empty retrievals or runaway tool/model spend without an obvious UI alarm.
• Scope: Frameworks components support agents; they are not the company ledger or the approval authority.