From dea93db8cc4a6a3852b156a1e76326ac0ad675a8 Mon Sep 17 00:00:00 2001 From: Tomas Beran Date: Thu, 19 Feb 2026 11:25:28 -0800 Subject: [PATCH] Add Coding Agents use case page (#120) --- docs.json | 1 + docs/use-cases/coding-agents.mdx | 57 ++++++++++++++++++++++++++++++++ 2 files changed, 58 insertions(+) create mode 100644 docs/use-cases/coding-agents.mdx diff --git a/docs.json b/docs.json index 2e722f14..03325f7f 100644 --- a/docs.json +++ b/docs.json @@ -45,6 +45,7 @@ { "group": "Use cases", "pages": [ + "docs/use-cases/coding-agents", "docs/use-cases/computer-use", "docs/use-cases/ci-cd" ] diff --git a/docs/use-cases/coding-agents.mdx b/docs/use-cases/coding-agents.mdx new file mode 100644 index 00000000..928dce82 --- /dev/null +++ b/docs/use-cases/coding-agents.mdx @@ -0,0 +1,57 @@ +--- +title: "Coding Agents" +description: "Run AI coding agents like Claude Code, Codex, and AMP in secure E2B sandboxes with full terminal, filesystem, and git access." +icon: "robot" +--- + +Coding agents like [Claude Code](https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/overview), [Codex](https://github.com/openai/codex), and [AMP](https://ampcode.com/) can write, debug, and refactor code autonomously. E2B sandboxes give each agent a full Linux environment with terminal, filesystem, and git — completely isolated from your infrastructure. Pre-built templates mean you can go from zero to a running agent in a single API call. + +## Why Use a Sandbox + +Running coding agents directly on your machine or servers means giving AI-generated code unrestricted access to your environment. E2B sandboxes solve this: + +1. **Isolation** — agent-generated code runs in a secure sandbox, never touching your production systems or local machine +2. **Full dev environment** — terminal, filesystem, git, and package managers are all available out of the box, so agents work like a developer would +3. **Pre-built templates** — ready-made templates for popular agents get you started fast, and you can [build your own](/docs/template/quickstart) for any agent +4. **Scalability** — spin up many sandboxes in parallel, each running its own agent on a separate task + +## How It Works + +1. **Create a sandbox** — use a pre-built template or [build your own](/docs/template/quickstart) with any agent installed +2. **Agent gets a full environment** — terminal, filesystem, git access, and any tools installed in the template +3. **Agent works autonomously** — it reads the codebase, writes code, runs tests, and iterates until the task is done +4. **Extract results** — pull out the git diff, structured output, or modified files via the SDK +5. **Sandbox is cleaned up** — once the work is done, the sandbox is destroyed automatically. No lingering state or cleanup needed + +## Agent Examples + +Since each sandbox is a full Linux environment, you can run any coding agent — just install it in a [custom template](/docs/template/quickstart). E2B also provides pre-built templates for popular agents to get you started quickly. + + + + Anthropic's autonomous coding agent with structured output and MCP tool support + + + OpenAI's coding agent with schema-validated output and image input + + + Sourcegraph's coding agent with streaming JSON and thread management + + + Open-source multi-provider agent with a built-in web UI + + + +## Related Guides + + + + Clone repos, manage branches, and push changes from sandboxes + + + Pause and resume sandboxes to preserve state across sessions + + + Build your own sandbox templates with custom tools and dependencies + +