AI Launch Profile

Claude Agent SDK opens Anthropic’s Claude Code agent infrastructure to developers

Anthropic introduced the Claude Agent SDK, described as the infrastructure used to build Claude Code and now available for developers building agents.

Read official source Tool profile

At a glance

Launch Snapshot

Company: Anthropic
Launch date: September 29, 2025
Launch type: New Product
Category: AI Agents, AI Developer Tools, AI Infrastructure
Audience: Developers who want to build agents with Claude Code-style memory, permission, and subagent patterns.
Pricing: Available through Anthropic developer tooling; costs depend on underlying Claude API/model usage.
Free plan: No
API: Yes
Open weights/source: No

Launch Context

Use these links to move from this record into the broader Launch Intelligence database.

Launch Intelligence hub Today's launches This week's launches More: AI Agents More: AI Developer Tools More: AI Infrastructure More: New Product

Verification & Sources

Evidence state: Recheck due
Source links: 2
Freshness: Needs recheck: checked June 8, 2026
Last updated: June 8, 2026

What this evidence state means

Definition: The claim was previously checked, but its review window expired or a material change may have invalidated it.
Required provenance: The prior evidence and check date are retained, together with the expiry or change signal that triggered recheck.
Owner: Kingy freshness queue owner and assigned editorial reviewer
Freshness rule: This is already outside its freshness rule. It must not be presented as current until reviewed against current evidence.
Disputes and corrections: Use “Suggest a correction” on the record. Kingy editorial reviews the cited evidence, records material corrections, and changes or removes the state when it is not supported.

Key source checks

Official site Claude developer platform

Suggest a correction

Kingy AI Take

Important because it productizes the agent infrastructure behind Claude Code rather than leaving teams to recreate orchestration patterns.

Who it is for

Developers who want to build agents with Claude Code-style memory, permission, and subagent patterns.

What feels promising

SDK-level access can help teams make more durable agent workflows with permissions and memory handled more deliberately.

What feels unproven

The SDK still requires careful design around tool permissions, prompt injection, data boundaries, and evaluation.

Source list

Sources

Official site www.anthropic.com Claude developer platform code.claude.com