[PR #441] [CLOSED] docs: add Tianshu integration guide #437

Closed
opened 2026-10-26 09:49:23 +00:00 by bsakaguchi · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-agent/pull/441
Author: @huiliyi37
Created: 10/6/2026
Status: ❌ Closed

Base: main ← Head: add-tianshu-harness


📝 Commits (2)

  • 32db1e9 Add a Tianshu guide for the current DeepSeek preset.
  • 3ff182b Use the Latin name Tianshu in the Chinese table.

📊 Changes

4 files changed (+150 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
📝 README.zh-CN.md (+1 -0)
➕ docs/tianshu.md (+74 -0)
➕ docs/tianshu.zh-CN.md (+74 -0)

📄 Description

Adds a bilingual integration guide for Tianshu and inserts it in alphabetical order in both README tables, between Reasonix and WorkBuddy.

This replaces the closed #287. That guide still installed tianshu-tui and described deepseek-v4-flash with reasoningEffort: max. The current preset is different:

  • Install tianshu-harness. tianshu and rivet are the same command.
  • Model ids are deepseek-flash (DeepSeek-V4.1-Flash, the default) and deepseek-v4-pro.
  • Both use a 1,000,000-token context window and a 256,000-token output cap.
  • Default reasoning_effort is high. /effort max requests max. A configured max is not stepped down.
  • deepseek-v4-flash was retired from the official preset on 2026-10-04. The guide says to configure deepseek-flash, and notes that Tianshu rewrites a saved official-provider entry off the old id.

tianshu config setup deepseek --key-env DEEPSEEK_API_KEY --default copies that preset. The first launch can also pick DeepSeek in /connect.

Disclosure: I maintain Tianshu.

Made with Cursor


🔄 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/441 **Author:** [@huiliyi37](https://github.com/huiliyi37) **Created:** 10/6/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `add-tianshu-harness` --- ### 📝 Commits (2) - [`32db1e9`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/32db1e97703d3058d758b71a08bb8203e1be2afa) Add a Tianshu guide for the current DeepSeek preset. - [`3ff182b`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/3ff182bd9f6be3f375d1477fe9d820a5fa26faa7) Use the Latin name Tianshu in the Chinese table. ### 📊 Changes **4 files changed** (+150 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) 📝 `README.zh-CN.md` (+1 -0) ➕ `docs/tianshu.md` (+74 -0) ➕ `docs/tianshu.zh-CN.md` (+74 -0) </details> ### 📄 Description Adds a bilingual integration guide for [Tianshu](https://github.com/huiliyi37/Tianshu-harness) and inserts it in alphabetical order in both README tables, between Reasonix and WorkBuddy. This replaces the closed #287. That guide still installed `tianshu-tui` and described `deepseek-v4-flash` with `reasoningEffort: max`. The current preset is different: - Install `tianshu-harness`. `tianshu` and `rivet` are the same command. - Model ids are `deepseek-flash` (DeepSeek-V4.1-Flash, the default) and `deepseek-v4-pro`. - Both use a 1,000,000-token context window and a 256,000-token output cap. - Default `reasoning_effort` is `high`. `/effort max` requests `max`. A configured `max` is not stepped down. - `deepseek-v4-flash` was retired from the official preset on 2026-10-04. The guide says to configure `deepseek-flash`, and notes that Tianshu rewrites a saved official-provider entry off the old id. `tianshu config setup deepseek --key-env DEEPSEEK_API_KEY --default` copies that preset. The first launch can also pick DeepSeek in `/connect`. Disclosure: I maintain Tianshu. Made with [Cursor](https://cursor.com) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
bsakaguchi 2026-10-26 09:49:23 +00:00
Sign in to join this conversation.
No labels
pull-request
No milestone
No project
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#437
No description provided.