# Claude Agent SDK

> SDK for building agents on the harness that powers Claude Code.

Claude Agent SDK has a SharkScore of 76.1/100, ranked #6 of 673 in Frameworks & SDKs (updated 2026-11-19).

- Website: https://code.claude.com/docs/en/agent-sdk/overview (code.claude.com)
- Kind: Tool · Pricing: Open source
- Maker: Anthropic
- Repository: https://github.com/anthropics/claude-agent-sdk-python
- Categories: Frameworks & SDKs, Agents & automation
- Verified by the maker: no

## Why it scores 76.1

| Part | Score (0–100) | Weight |
|---|---|---|
| Adoption | 67 | 30% |
| User reviews | no data | 25% |
| Maintenance | 100 | 20% |
| Trust & transparency | 80 | 15% |
| Momentum | no data | 10% |

Evidence covers 87% of the score's weight. Compared with other listings in Frameworks & SDKs.

## Evidence

| Signal | Value |
|---|---|
| Traffic rank (Tranco) | — |
| GitHub stars | 8.6K |
| Stars, last 90 days | — |
| Downloads, 30 days | 111M |
| Reviews | none yet |
| Last update | 0 days ago |
| Uptime, 30 days | 100.0% |
| Public pricing / Free tier / API / MCP | yes / yes / no / yes |

## RankShark review

Automated evaluation by RankShark · 2026-11-18 · method review-1.0 · https://rankshark.pro/how-we-review
We were not paid for this review.

As of 2026-11-18, Claude Agent SDK is a Python and TypeScript library for building agents with Claude Code’s tools and agent loop; API-key authentication is required for developers building products (E1, E3).

The vendor describes Claude Agent SDK as a library for embedding Claude Code’s agent in Python or TypeScript applications. It provides built-in tools, permissions, sessions, hooks, and context management. (E1)

### Pros

- The SDK includes an agent loop and context management rather than requiring developers to build those parts from scratch.
- The documented capabilities include tools, permissions, hooks, sessions, subagents, MCP, and plugins.
- The vendor documents Python and TypeScript SDKs and provides a CLI subprocess option for other languages.

### Cons

- The SDK is documented for Python and TypeScript; other languages use the CLI as a subprocess rather than a native SDK.
- Developers building products that use Claude capabilities are directed to API-key authentication; offering Claude.ai login or subscription rate limits requires prior approval.
- No public prices found on the pages we checked.
- No free plan or free trial stated on the pages we checked.
- Whether customer data is used for model training is not stated on the pages we checked.

### Pricing

No public prices found on the pages we checked (2026-11-18).

### Sources

- https://code.claude.com/docs/en/agent-sdk/overview (fetched 2026-11-18)
- https://code.claude.com/docs/llms.txt (fetched 2026-11-18)
- https://code.claude.com/docs/en/legal-and-compliance (fetched 2026-11-18)
- https://code.claude.com/docs/en/legal-and-compliance (fetched 2026-11-18)

## Top alternatives

1. [Model Context Protocol](https://rankshark.pro/tools/model-context-protocol) — SharkScore 81.7. Open protocol and SDKs that connect AI apps to tools and data.
2. [AI SDK](https://rankshark.pro/tools/ai-sdk) — SharkScore 77.7. TypeScript toolkit for building AI apps and agents.
3. [LangChain](https://rankshark.pro/tools/langchain) — SharkScore 77.7. Build agents and LLM apps in Python and TypeScript.
4. [LangSmith](https://rankshark.pro/tools/langsmith) — SharkScore 76.9. Tracing, evaluation and monitoring for LLM apps.
5. [Transformers](https://rankshark.pro/tools/transformers) — SharkScore 76.9. Hugging Face's library of pretrained models for text, vision and audio.

---
Source: https://rankshark.pro/tools/claude-agent-sdk · Method v2026-11-18.2: https://rankshark.pro/how-ranking-works · Payment never changes a rank.
