[PR #626] feat: add deepseek-kit to AI Agent frameworks #633

New issue

Open

opened 2026-10-26 10:00:05 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 10:00:05 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-integration/pull/626
Author: @FliPPeDround
Created: 5/21/2026
Status: 🔄 Open

Base: main ← Head: add-deepseek-kit

📝 Commits (1)

109de2e feat: add deepseek-kit to AI Agent frameworks

📊 Changes

8 files changed (+185 additions, -0 deletions)

View changed files

📝 README.md (+5 -0)
📝 README_cn.md (+5 -0)
📝 README_es.md (+5 -0)
📝 README_ja.md (+5 -0)
📝 README_zh_tw.md (+5 -0)
➕ docs/deepseek-kit/README.md (+80 -0)
➕ docs/deepseek-kit/README_cn.md (+80 -0)
➕ docs/deepseek-kit/assets/logo.png (+0 -0)

📄 Description

Add deepseek-kit to AI Agent frameworks

Project Information

Name: deepseek-kit
Repository: https://github.com/FliPPeDround/deepseek-kit
npm: deepseek-kit
Documentation: https://deepseek-kit.netlify.app
License: MIT

Description

deepseek-kit is a lightweight TypeScript Agent framework with native-level DeepSeek adaptation, solving critical issues that general-purpose frameworks like LangChain.js and AI SDK cannot properly handle:

🧠 Precise Tool Calling in Thinking Mode — Automatically tracks and re-sends reasoning_content in the agent loop, applies differentiated strategies based on whether tool calls occurred, and enables thinking mode by default — zero configuration needed.
💾 Maximum Cache Hit Rate — Sends zero-redundancy request bodies with deterministic message construction, ensuring the same input always produces the same request prefix for optimal DeepSeek API cache utilization.
📋 Reliable Structured Output — Zod Schema-driven structured output with smart retry and formatted error feedback, fully compatible with thinking mode.

Changes

Added docs/deepseek-kit/README.md (English documentation)
Added docs/deepseek-kit/README_cn.md (Chinese documentation)
Added docs/deepseek-kit/assets/logo.png
Added deepseek-kit entry to AI Agent frameworks section in README.md
Added deepseek-kit entry to AI Agent 框架 section in README_cn.md
Added deepseek-kit entry to AI Agent frameworks section in README_zh_tw.md
Added deepseek-kit entry to AI エージェントフレームワーク section in README_ja.md
Added deepseek-kit entry to Marcos de agentes de IA section in README_es.md

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/awesome-deepseek-integration/pull/626 **Author:** [@FliPPeDround](https://github.com/FliPPeDround) **Created:** 5/21/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `add-deepseek-kit` --- ### 📝 Commits (1) - [`109de2e`](https://github.com/deepseek-ai/awesome-deepseek-integration/commit/109de2ec25b8e7e5069cca016a638a765a569f50) feat: add deepseek-kit to AI Agent frameworks ### 📊 Changes **8 files changed** (+185 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+5 -0) 📝 `README_cn.md` (+5 -0) 📝 `README_es.md` (+5 -0) 📝 `README_ja.md` (+5 -0) 📝 `README_zh_tw.md` (+5 -0) ➕ `docs/deepseek-kit/README.md` (+80 -0) ➕ `docs/deepseek-kit/README_cn.md` (+80 -0) ➕ `docs/deepseek-kit/assets/logo.png` (+0 -0) </details> ### 📄 Description ## Add deepseek-kit to AI Agent frameworks ### Project Information - **Name**: deepseek-kit - **Repository**: https://github.com/FliPPeDround/deepseek-kit - **npm**: [deepseek-kit](https://npmx.dev/package/deepseek-kit) - **Documentation**: https://deepseek-kit.netlify.app - **License**: MIT ### Description deepseek-kit is a lightweight TypeScript Agent framework with native-level DeepSeek adaptation, solving critical issues that general-purpose frameworks like LangChain.js and AI SDK cannot properly handle: - 🧠 **Precise Tool Calling in Thinking Mode** — Automatically tracks and re-sends `reasoning_content` in the agent loop, applies differentiated strategies based on whether tool calls occurred, and enables thinking mode by default — zero configuration needed. - 💾 **Maximum Cache Hit Rate** — Sends zero-redundancy request bodies with deterministic message construction, ensuring the same input always produces the same request prefix for optimal DeepSeek API cache utilization. - 📋 **Reliable Structured Output** — Zod Schema-driven structured output with smart retry and formatted error feedback, fully compatible with thinking mode. ### Changes - Added `docs/deepseek-kit/README.md` (English documentation) - Added `docs/deepseek-kit/README_cn.md` (Chinese documentation) - Added `docs/deepseek-kit/assets/logo.png` - Added deepseek-kit entry to AI Agent frameworks section in `README.md` - Added deepseek-kit entry to AI Agent 框架 section in `README_cn.md` - Added deepseek-kit entry to AI Agent frameworks section in `README_zh_tw.md` - Added deepseek-kit entry to AI エージェントフレームワーク section in `README_ja.md` - Added deepseek-kit entry to Marcos de agentes de IA section in `README_es.md` --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi added the

pull-request

label

2026-10-26 10:00:05 +00:00

bsakaguchi referenced this issue

2026-10-26 10:00:10 +00:00

[PR #633] feat: add llm-api-translator to AI Agent frameworks #641