chore: docs

Author: NathanFlurry

README.md

@@ -208,7 +208,7 @@ Yes. For orchestrating stateful, long-running tasks, we recommend pairing Secure
 <details>
 <summary>What are common use cases?</summary>
 
-- [AI agent code evaluation and tool use](https://secureexec.dev/docs/use-cases/ai-agent-code-eval)
+- [AI agent code execution and tool use](https://secureexec.dev/docs/use-cases/ai-agent-code-exec)
 - [User-facing dev servers (Express, Hono, Next.js)](https://secureexec.dev/docs/use-cases/dev-servers)
 - MCP tool-code execution
 - [Sandboxed plugin / extension systems](https://secureexec.dev/docs/use-cases/plugin-systems)

docs/docs.json

@@ -39,7 +39,8 @@
       {
         "group": "Use Cases",
         "pages": [
-          "use-cases/ai-agent-code-eval",
+          "use-cases/ai-agent-code-exec",
+          "use-cases/code-mode",
           "use-cases/dev-servers",
           "use-cases/plugin-systems"
         ]

docs/features/child-processes.mdx

@@ -4,6 +4,10 @@ description: Spawn child processes from sandboxed code.
 icon: "terminal"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/child-processes.ts">
+  Runnable example for child process spawning.
+</Card>
+
 Sandboxed code can spawn child processes through the `CommandExecutor` interface, gated by the `childProcess` permission.
 
 ## Runnable example

docs/features/filesystem.mdx

@@ -4,6 +4,10 @@ description: Filesystem backends for sandboxed code.
 icon: "folder"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/filesystem.ts">
+  Runnable example for filesystem backends.
+</Card>
+
 secure-exec supports three filesystem backends. The system driver controls which backend sandboxed code uses.
 
 ## Runnable example

docs/features/module-loading.mdx

@@ -5,6 +5,10 @@ description: How sandboxed code resolves and loads modules.
 icon: "cubes"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/module-loading.ts">
+  Runnable example for module resolution and loading.
+</Card>
+
 Sandboxed code can `require()` and `import` modules through secure-exec's module resolution system.
 
 ## Runnable example

docs/features/networking.mdx

@@ -4,6 +4,10 @@ description: Network access for sandboxed code.
 icon: "globe"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/networking.ts">
+  Runnable example for network access control.
+</Card>
+
 Network access is deny-by-default. Enable it by setting `useDefaultNetwork: true` on the system driver and granting the `network` permission.
 
 ## Runnable example

docs/features/output-capture.mdx

@@ -4,6 +4,10 @@ description: Capture console output from sandboxed code.
 icon: "message-lines"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/output-capture.ts">
+  Runnable example for capturing console output.
+</Card>
+
 Console output from sandboxed code is **not buffered** into result fields. `exec()` and `run()` do not return `stdout` or `stderr`. Use the `onStdio` hook to capture output.
 
 ## Runnable example

docs/features/permissions.mdx

@@ -4,6 +4,10 @@ description: Control what sandboxed code can access on the host.
 icon: "lock"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/permissions.ts">
+  Runnable example for permission configuration.
+</Card>
+
 All host capabilities are **deny-by-default**. Sandboxed code cannot access the filesystem, network, child processes, or environment variables unless you explicitly allow it.
 
 ## Runnable example

docs/features/resource-limits.mdx

@@ -4,6 +4,10 @@ description: CPU time, memory, and payload size controls.
 icon: "gauge"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/resource-limits.ts">
+  Runnable example for CPU and memory limits.
+</Card>
+
 Resource limits prevent sandboxed code from running forever or exhausting host memory.
 
 ## Runnable example

docs/features/typescript.mdx

@@ -4,6 +4,10 @@ description: Sandboxed type checking and compilation.
 icon: "code"
 ---
 
+<Card title="Example on GitHub" icon="github" href="https://github.com/anthropics/secure-exec/blob/main/examples/features/src/typescript.ts">
+  Runnable example for sandboxed TypeScript compilation.
+</Card>
+
 The `@secure-exec/typescript` companion package runs the TypeScript compiler inside a sandbox for safe type checking and compilation of untrusted code.
 
 ## Runnable example