Claude Agent SDK
Anthropic
Formerly Claude Code SDK
Anthropic's Python/TypeScript library that embeds Claude Code's agent loop, tools, hooks, subagents and MCP in your own app.
Curated by Tiberiu ArvaVerified
Strengths
- Claude Code tools, hooks and subagents
- Permission modes for tool approval
- Bedrock, Vertex and Foundry auth options
Practitioner note
Fastest path to a Claude Code-grade agent inside your own process, but it is Claude-only and usage is billed through the Anthropic API or the Bedrock/Vertex/Foundry route you configure. Anthropic does not allow third-party products to offer claude.ai login, so plan for API-key auth.
Warning
Governance posture
Reviewed 2026-09-30Library runs in the operator's process; prompts go to the chosen provider (Anthropic API, Amazon Bedrock, Google Cloud, Microsoft Foundry), whose residency terms apply.
SourceLibrary embedded in an application you operate; Anthropic-hosted Managed Agents is a separate product. (self-hosted)
SourceNo built-in audit log; hooks let the operator intercept and log tool calls and lifecycle events.
SourceSelf-hosted library; SOC 2 applies to the operator and to the chosen model provider, not the package.
Self-hosted library; ISO 27001 applies to the operator and to the chosen model provider, not the package.
Self-hosted library; ISO 42001 applies to the deploying organisation and model provider.
Developer component, not an AI system placed on the market; obligations rest with the deployer (GPAI model obligations sit with Anthropic). (role: not-applicable)
MIT wrapper, but bundled Claude Code binary and Anthropic Commercial Terms govern use; tied to Claude models. (medium)
SourceEU AI Act obligations
Risk tier: Not applicable · as of 2026-08-23Source-backed information mapped from Claude Agent SDK's tracked risk tier — not legal advice. Obligations depend on how your organisation deploys the system; see the full obligation reference and timeline.
This record's EU AI Act risk tier is marked not applicable, so no tier-specific obligations attach. Deployers embedding it in an AI system in scope of the Act should assess that system's own tier.
Explore the category
Compare this tool against the rest of its category and the cloud platform foundation layer.
Return to the category grid to compare governance posture across every tracked tool.
Review Microsoft Foundry, Amazon Bedrock, and Gemini Enterprise Agent Platform as the foundation layer.
Track releases, deprecations, license changes, and other market movements.