# deepseek-ai/dsh-superagent-prototype

> Research prototype of a proactive coding agent for a single GitHub repository, packaged as a bundle for DeepSeek Harness: it listens to GitHub and Feishu, opens work items and replies in the original conversation, for example with a pull request.

- Magnitude: 8.8 out of 10 — Early signal
- Stars: 1,299 total · +1,032 stars measured 2026-10-31, 13:38–16:24 UTC, ≈ 1,350 by evening
- Star trust: star growth looks organic
- Category: AI agents · Language: TypeScript · License: MIT · Created: 2026-10-31 · Last push: 2026-10-31
- GitHub: https://github.com/deepseek-ai/dsh-superagent-prototype · Page: https://gitnova.dev/en/r/deepseek-ai/dsh-superagent-prototype

## Useful for

- Connect the agent to a repository so it opens PRs from GitHub requests
- Define rules that let the agent act without an explicit team request
- Check the web panel to see which agent tasks need a human

## Why it’s here

- Star-counter measurements on 2026-10-31 (UTC), 13:38–16:24: 267 → 1299 stars (+1032). This is the change over that interval.
- Estimated end-of-day forecast: about +1,350 stars, using observed gains and the previous day.
- The repository is 0 days old and already has 1,299 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.
- Top new repositories this week: #15.
- Hacker News: “DeepSeek open-sources dsh-superagent, a coding agent largely built by agents” — 7 points, 2 h ago.
- About 280 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 97
- Issues and pull requests: 0
- Watchers: 7
- Average over the last week: 1,350 per day
- Usual pace: too little history (under two weeks)
- Stars in the last hour (measured): 384

## Stars per day, last 1 day (oldest → newest, today is partial)

2026-10-31 … 2026-10-31: 1032

## Hacker News

- DeepSeek open-sources dsh-superagent, a coding agent largely built by agents — 7 points, 0 comments: https://news.ycombinator.com/item?id=50316116

## Spotted in now

- Top new repositories this week: #15

## More in this category

1. **wangzan101/Mika** — 8.7 · Breakout · AI agents · Kotlin · +1,179 stars measured 2026-10-31, 00:27–16:24 UTC, ≈ 1,692 by evening
   Mika is an on-device AI agent for Android that carries out one-sentence requests by operating apps and the web on a hidden background screen while you keep using your phone. It uses your own model API key, with no project servers.
   Full card: https://gitnova.dev/en/r/wangzan101/Mika.md
2. **pingdotgg/t3code** — 8.5 · Breakout · AI agents · TypeScript · +1,644 stars measured 2026-10-31, 00:27–16:24 UTC, ≈ 2,245 by evening
   T3 Code is an agent harness control surface for agents on your machine, with mobile (iOS, Android), web and Electron desktop apps. It works with your Claude Code, Codex, Cursor, Grok Build, OpenCode and Google Antigravity subscriptions if…
   Full card: https://gitnova.dev/en/r/pingdotgg/t3code.md
3. **ccch1mneyyy/dsh-Go** — 8.5 · Breakout · AI agents · Dart · +968 stars measured 2026-10-31, 00:27–16:24 UTC, ≈ 1,457 by evening
   Unofficial Android client for DeepSeek Harness: runs DSH on the phone in a built-in Linux environment without root or Termux, and can also remotely control DSH on your computer.
   Full card: https://gitnova.dev/en/r/ccch1mneyyy/dsh-Go.md
4. **milind-soni/OpenMausBot** — 8.3 · Breakout · AI agents · TypeScript · +705 stars measured 2026-10-31, 00:27–16:24 UTC, ≈ 1,041 by evening
   ,
   Full card: https://gitnova.dev/en/r/milind-soni/OpenMausBot.md
5. **heydamir/unslop** — 8.2 · Breakout · AI agents · JavaScript · +1,638 stars measured 2026-10-31, 00:27–16:24 UTC, ≈ 2,300 by evening
   A Claude skill that audits and redesigns AI-generated, «vibecoded» UI screen by screen, replacing purple gradients, glows and emoji labels with a deliberate design direction.
   Full card: https://gitnova.dev/en/r/heydamir/unslop.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-31 16:35 UTC, updated every 30 minutes.