agent-sandbox-skilldisler/agent-sandbox-skill
Provides isolated sandbox execution environments for AI agents, so multiple agent forks can run code, drive browsers,…
No ratings yet
Rate this skill
Stars388
Forks97
Watchers388
Open Issues2
InstallOne-line install
UnconfirmedWhether this repository follows the Agent Skills spec (that is, whether it contains a SKILL.md) has not been verified yet. Until that is settled, clone it and use what you need:
git clone https://github.com/disler/agent-sandbox-skill.git
OverviewOverview
Key Features
Provides isolated sandbox execution environments for AI agents, so multiple agent forks can run code, drive browsers, and do full-stack development in parallel without interfering with each other. Key capabilities include: 1. Isolated Execution: each agent fork runs in a fully isolated sandbox, so one fork's code, dependencies, and processes never pollute other forks or the host machine; 2. Elastic Scale: spin up as many agent forks as needed to work in parallel, with task scale not limited by a single machine's resources; 3. Browser Automation: drive a browser directly inside the sandbox to perform page interactions and data scraping without configuring a local browser environment; 4. Full-Stack Development Support: the sandbox ships with toolchains such as uv, letting agents clone repositories, install dependencies, and run front-end and back-end projects end to end; 5. Persistent Context: agent forks retain their working directory and runtime state across interactions, so long tasks do not need to rebuild the environment repeatedly.
Use Cases
isolating each agent's changes when multiple agents develop the same project in parallel, validating that generated code actually runs in a clean environment, letting agents autonomously complete end-to-end tasks that require a browser, quickly bootstrapping full-stack project prototypes, and long-running automation tasks that need state preserved across turns.
Target Roles
AI / Algorithm Engineer, Full-Stack / Indie Developer, Test / QA Engineer, DevOps / Operations Engineer.
RatingUser ratings
Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.
RepositoryRepository
| Skill ID | 2062 |
|---|---|
| Data source | GitHub |
| Repository | disler/agent-sandbox-skill |
| Category | AI & Agent Skills / Agent Runtime & Sandbox |
| Primary language | Python |
| License | Not specified |
| Default branch | main |
| GitHub repository ID | 1102089336 |
| Node ID | R_kgDOQbCMeA |
| Rank | #1571 (position within the source ranking) |
| Site-wide popularity rank | #1774 (computed live from Stars, across all 5833 entries) |
GrowthStar / Fork growth
One snapshot per day, tracking how Stars and Forks changed
- Now388
- Previous point · 09-18 387
- vs previous point +1
- 2 data points 23 days
- Now97
- Previous point · 09-18 97
- vs previous point ±0
- 2 data points 23 days
TimelineTimeline
Repository created2025-11-22 19:53:0212 months ago
Data last updated2026-10-26 12:16:2823 days ago
Last commit2025-11-29 18:57:4611 months ago
CommitLatest commit
Version [latest] committed by disler. Update Category: Codebase Iteration & Routine Maintenance. Commit details: browser tools. This update updates core functional modules and advances project iteration.