docs: add SDK examples for updateNetwork in proxy configuration

[devin-ai-integration]

Summary

Updates the "Update proxy configuration" section in Sandboxes/Proxy.mdx to showcase the new SDK methods:

• TypeScript: SandboxInstance.updateNetwork(sandboxName, { network: {...} })
• Python: SandboxInstance.update_network(sandbox_name, SandboxUpdateNetwork(network={...}))

The existing cURL example is preserved and wrapped in a <CodeGroup> alongside the new SDK examples.

Refs: ENG-2665

Review & Testing Checklist for Human

• Verify the TypeScript example matches the actual updateNetwork method signature from sdk-typescript#313
• Verify the Python example matches the actual update_network method signature from sdk-python#150
• Preview the rendered MDX page to ensure <CodeGroup> tabs display correctly

Notes

The SDK methods were added in sdk-typescript#313 and sdk-python#150.

Link to Devin session: https://app.devin.ai/sessions/0f7a2e36861d46ecb6c11d4749803173
Requested by: @Joffref

---

Note

Adds TypeScript and Python SDK examples for updateNetwork/update_network to the "Update proxy configuration" section, wrapping all three examples (TypeScript, Python, cURL) in a <CodeGroup> component.

^{Written by Mendral for commit 7640167.}

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR that start with 'DevinAI' or '@devin'.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring

[mendral-app]

Needs attention — 1 issue in 1 file

The Python example passes a raw dict with camelCase keys ("allowedDomains", "routing") to SandboxUpdateNetwork(network={...}). If the SDK's Pydantic model uses snake_case field names (the Python convention), this will silently ignore or reject those keys at runtime — readers copying this example verbatim will get an empty/incorrect config. The PR checklist correctly flags signature verification as required before merge; that check must confirm the dict key casing is actually valid for the Python SDK.

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.

<assessment>
The Python example passes a raw `dict` with camelCase keys (`"allowedDomains"`, `"routing"`) to `SandboxUpdateNetwork(network={...})`. If the SDK's Pydantic model uses snake_case field names (the Python convention), this will silently ignore or reject those keys at runtime — readers copying this example verbatim will get an empty/incorrect config. The PR checklist correctly flags signature verification as required before merge; that check must confirm the dict key casing is actually valid for the Python SDK.
</assessment>

<file name="Sandboxes/Proxy.mdx">
<issue location="Sandboxes/Proxy.mdx:251">
Python dict keys use camelCase (`allowedDomains`, `routing`) where the Python SDK likely expects snake_case (`allowed_domains`). If `SandboxUpdateNetwork` is a Pydantic model without camelCase aliases, these keys will be silently dropped, producing an empty network config.
</issue>
</file>

_{Tag @mendral-app with feedback or questions. View session}

[mendral-app]

bug (P2): Python dict keys use camelCase (allowedDomains, routing) where the Python SDK likely expects snake_case (allowed_domains). If SandboxUpdateNetwork is a Pydantic model without camelCase aliases, these keys will be silently dropped, producing an empty network config.

Suggested change

Suggested change

	SandboxUpdateNetwork(network={
	"allowedDomains": ["api.stripe.com", "api.openai.com", "api.anthropic.com"],
	"proxy": {
	"routing": [
	{
	"destinations": ["api.stripe.com"],
	"headers": {"Authorization": "Bearer {{SECRET:stripe-key}}"},
	"secrets": {"stripe-key": "sk_live_REPLACE_WITH_YOUR_KEY"},
	},
	{
	"destinations": ["api.anthropic.com"],
	"headers": {"x-api-key": "{{SECRET:anthropic-key}}"},
	"secrets": {"anthropic-key": "REPLACE_WITH_YOUR_ANTHROPIC_KEY"},
	},
	],
	},
	}),
	)
	await SandboxInstance.update_network(
	"my-sandbox",
	SandboxUpdateNetwork(network={
	"allowed_domains": ["api.stripe.com", "api.openai.com", "api.anthropic.com"],
	"proxy": {
	"routing": [
	{
	"destinations": ["api.stripe.com"],
	"headers": {"Authorization": "Bearer {{SECRET:stripe-key}}"},
	"secrets": {"stripe-key": "sk_live_REPLACE_WITH_YOUR_KEY"},
	},
	{
	"destinations": ["api.anthropic.com"],
	"headers": {"x-api-key": "{{SECRET:anthropic-key}}"},
	"secrets": {"anthropic-key": "REPLACE_WITH_YOUR_ANTHROPIC_KEY"},
	},
	],
	},
	}),
	)

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At Sandboxes/Proxy.mdx, line 251:

<issue>
Python dict keys use camelCase (`allowedDomains`, `routing`) where the Python SDK likely expects snake_case (`allowed_domains`). If `SandboxUpdateNetwork` is a Pydantic model without camelCase aliases, these keys will be silently dropped, producing an empty network config.
</issue>

[devin-ai-integration]

Not a bug — the Python SDK integration tests use camelCase dict keys:

SandboxUpdateNetwork(network={"allowedDomains": ["httpbin.org", "*.httpbin.org"]})
SandboxUpdateNetwork(network={"forbiddenDomains": ["httpbin.org", "*.httpbin.org"]})

See tests/integration/core/sandbox/test_network.py. The network parameter accepts a Union[SandboxNetwork, Dict[str, Any]] — when a dict is passed, it goes through the API client which uses camelCase field names matching the REST API schema.