deepseek-ai/awesome-deepseek-agent
1
0
Fork 0
mirror of https://github.com/deepseek-ai/awesome-deepseek-agent.git synced 2026-11-05 01:07:37 +00:00
Code Issues 354 Projects Packages Wiki Activity

[PR #143] Add OpenClacky integration guide #228

New issue
Open
opened 2026-10-26 09:47:03 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-26 09:47:03 +00:00
Owner
Copy link

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-agent/pull/143
Author: @windy
Created: 5/23/2026
Status: 🔄 Open

Base: main ← Head: add-openclacky

📝 Commits (1)

  • 05554c6 Add OpenClacky integration guide

📊 Changes

4 files changed (+164 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
📝 README.zh-CN.md (+1 -0)
➕ docs/openclacky.md (+81 -0)
➕ docs/openclacky.zh-CN.md (+81 -0)

📄 Description

Summary

Adds an integration guide for OpenClacky — an MIT-licensed open-source AI agent with native DeepSeek support,
~100% prompt-cache hit rate, and a Web UI / Terminal / IM (Feishu, WeCom, WeChat, Discord, Telegram) experience from one binary.

Changes

  • docs/openclacky.md — English guide
  • docs/openclacky.zh-CN.md — Simplified Chinese guide
  • README.md / README.zh-CN.md — table entry inserted in alphabetical order (before OpenClaw)

Checklist

Agent Configuration

  • Model names use the current deepseek-v4-pro / deepseek-v4-flash (no deepseek-chat / deepseek-reasoner / deepseek-coder)
  • 1M context window documented (context_window: 1000000)
  • reasoning_effort=max recommended; guide explicitly notes OpenClacky handles reasoning_content correctly so users do not need to disable
    thinking mode
  • No workaround for already-fixed upstream bugs — DeepSeek V4 thinking-mode support is native
  • No pricing table included (so no stale numbers to maintain)
  • Configuration fields shown (/config → Provider / Base URL / API Key / Model / Context window / Reasoning effort) are verified to work in the agent

Documentation

  • Both English and Chinese versions in a single PR
  • README table entry added in alphabetical order in both README.md and README.zh-CN.md
  • Single tool per PR
  • No images added (no docs/assets/ changes)

Notes for reviewers

OpenClacky's value prop for DeepSeek users is specifically token efficiency:

  • session never restarts and the system prompt is never mutated → near-100% prompt cache hit, which compounds with DeepSeek's $0.003625 / M tokens
    cache-hit price
  • 16 core tools + a single invoke_skill meta-tool → small request payload leaves more room in the 1M context
  • BYOK routing → main agent on deepseek-v4-pro, subagents fan out to deepseek-v4-flash

🔄 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-agent/pull/143 **Author:** [@windy](https://github.com/windy) **Created:** 5/23/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `add-openclacky` --- ### 📝 Commits (1) - [`05554c6`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/05554c6a85baa950e9cacdbcfc4d5397df033fd4) Add OpenClacky integration guide ### 📊 Changes **4 files changed** (+164 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) 📝 `README.zh-CN.md` (+1 -0) ➕ `docs/openclacky.md` (+81 -0) ➕ `docs/openclacky.zh-CN.md` (+81 -0) </details> ### 📄 Description ## Summary Adds an integration guide for [OpenClacky](https://github.com/clacky-ai/openclacky) — an MIT-licensed open-source AI agent with native DeepSeek support, ~100% prompt-cache hit rate, and a Web UI / Terminal / IM (Feishu, WeCom, WeChat, Discord, Telegram) experience from one binary. ## Changes - `docs/openclacky.md` — English guide - `docs/openclacky.zh-CN.md` — Simplified Chinese guide - `README.md` / `README.zh-CN.md` — table entry inserted in alphabetical order (before `OpenClaw`) ## Checklist ### Agent Configuration - [x] Model names use the current `deepseek-v4-pro` / `deepseek-v4-flash` (no `deepseek-chat` / `deepseek-reasoner` / `deepseek-coder`) - [x] 1M context window documented (`context_window: 1000000`) - [x] `reasoning_effort=max` recommended; guide explicitly notes OpenClacky handles `reasoning_content` correctly so users do **not** need to disable thinking mode - [x] No workaround for already-fixed upstream bugs — DeepSeek V4 thinking-mode support is native - [x] No pricing table included (so no stale numbers to maintain) - [x] Configuration fields shown (`/config` → Provider / Base URL / API Key / Model / Context window / Reasoning effort) are verified to work in the agent ### Documentation - [x] Both English and Chinese versions in a single PR - [x] README table entry added in alphabetical order in both `README.md` and `README.zh-CN.md` - [x] Single tool per PR - [x] No images added (no `docs/assets/` changes) ## Notes for reviewers OpenClacky's value prop for DeepSeek users is specifically token efficiency: - session never restarts and the system prompt is never mutated → near-100% prompt cache hit, which compounds with DeepSeek's $0.003625 / M tokens cache-hit price - 16 core tools + a single `invoke_skill` meta-tool → small request payload leaves more room in the 1M context - BYOK routing → main agent on `deepseek-v4-pro`, subagents fan out to `deepseek-v4-flash` --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/awesome-deepseek-agent#228
No description provided.