# mattpocock/sandcastle

> TypeScript library for running AI coding agents in isolated sandboxes: a single sandcastle.run() call creates a branch, runs the agent, and merges its commits back. Supports native (Bubblewrap, Seatbelt), container (Docker, Podman), and cloud (E2B, Modal, Vercel, etc.) providers.

- Magnitude: 1.9 out of 10 — Steady
- Stars: 8,805 total · +3 stars measured 2026-11-03, 11:25–13:57 UTC, ≈ 8 by evening
- Star trust: star growth looks organic
- Category: AI agents · Language: TypeScript · License: MIT · Created: 2026-03-17 · Last push: 2026-11-03
- GitHub: https://github.com/mattpocock/sandcastle · Page: https://gitnova.dev/en/r/mattpocock/sandcastle

## Useful for

- Run multiple AFK agents in parallel across isolated sandboxes
- Build a review pipeline where an agent edits code on an isolated branch
- Plug in a custom sandbox provider via createSandbox()

## Why it’s here

- Star-counter measurements on 2026-11-03 (UTC), 11:25–13:57: 8802 → 8805 stars (+3). This is the change over that interval.
- Estimated end-of-day forecast: about +8 stars, using observed gains and the previous day.
- GitHub Trending TypeScript today: #12, +28 stars.

## 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: 942
- Issues and pull requests: 1,526
- Watchers: 24
- Average over the last week: 20 per day
- Usual pace: 17 per day
- Stars in the last hour (measured): 1
- Latest release: v0.14.0 (2026-11-02)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-10-05 … 2026-11-03: 22, 15, 9, 23, 19, 22, 14, 28, 24, 17, 24, 17, 22, 16, 15, 20, 23, 28, 15, 11, 15, 12, 17, 27, 18, 16, 18, 31, 23, 3

## Spotted in now

- GitHub Trending TypeScript today: #12, +28 stars

## More in this category

1. **storytold/craft-anything** — 9.3 · Breakout · AI agents · Rust · +2,335 stars measured 2026-11-03, 00:30–13:47 UTC, ≈ 3,907 by evening
   A pure-Rust reverse-engineering engine with 284 tools that inspects websites, apps and games, then has a coding agent rebuild a similar version and checks the result against the original.
   Full card: https://gitnova.dev/en/r/storytold/craft-anything.md
2. **vercel-labs/agent-desktop** — 9.2 · Breakout · AI agents · Rust · +2,097 stars measured 2026-11-03, 00:30–13:47 UTC, ≈ 3,557 by evening
   Rust CLI that lets AI agents control desktop apps via macOS accessibility trees, with stable element refs and a snapshot-to-action loop. Combines a native binary with a C-ABI cdylib for Python/Swift/Go/Node bindings.
   Full card: https://gitnova.dev/en/r/vercel-labs/agent-desktop.md
3. **pingdotgg/t3code** — 9.0 · Breakout · AI agents · TypeScript · +1,421 stars measured 2026-11-03, 00:30–13:47 UTC, ≈ 2,640 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
4. **deepseek-ai/dsh-superagent-prototype** — 8.6 · Peaking · AI agents · TypeScript · +2,631 stars measured 2026-11-03, 00:30–13:47 UTC, ≈ 4,508 by evening
   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…
   Full card: https://gitnova.dev/en/r/deepseek-ai/dsh-superagent-prototype.md
5. **ccch1mneyyy/dsh-Go** — 8.0 · Breakout · AI agents · Dart · +1,022 stars measured 2026-11-03, 00:30–13:47 UTC, ≈ 1,641 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

---

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 2026-11-03 14:05 UTC, updated every 30 minutes.