[Agents] Document MCP and Code Mode SDK updates

[mattzcarey]

Summary

Documents two Agents SDK updates in one changelog entry:

• MCP schema conversion caching and Think.includeMcpTools from cloudflare/agents#1959
• framework-neutral execute(), search(), and describe() Code Mode runtime APIs from cloudflare/agents#1969, available in @cloudflare/codemode@latest

Updates the Think, MCP client, Code Mode runtime, and Code Mode API reference pages. Search and describe documentation also covers the new requiresApproval result metadata.

Fixes the documentation gap tracked by cloudflare/agents#1938.

Documentation checklist

• Is there a changelog entry (guidelines)? If you don't add one for something awesome and new (however small) — how will our customers find out? Changelogs are automatically posted to RSS feeds, the Discord, and X.
• The change adheres to the documentation style guide.

[github-actions]

This pull request requires reviews from CODEOWNERS as it changes files that match the following patterns:

Pattern	Owners
/src/content/changelog/	@cloudflare/pm-changelogs, @cloudflare/product-owners
/src/content/docs/agents/	@irvinebroque, @rita3ko, @elithrar, @cloudflare/product-owners, @cloudflare/ai-agents, @cloudflare/dev-plat-leads

[github-actions]

Nimbus Preview URL: https://nimbus-ca9c15c6.preview.developers.cloudflare.com
Nimbus Preview Branch URL: https://nimbus-docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com

[github-actions]

Preview URL: https://180732ab.preview.developers.cloudflare.com
Preview Branch URL: https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com

Files with changes (up to 15)

Original Link	Updated Link
https://developers.cloudflare.com/changelog/post/2026-07-22-mcp-codemode-updates/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/changelog/post/2026-07-22-mcp-codemode-updates/
https://developers.cloudflare.com/agents/tools/codemode/durable-runtime/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/tools/codemode/durable-runtime/
https://developers.cloudflare.com/agents/harnesses/think/tools/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/harnesses/think/tools/
https://developers.cloudflare.com/agents/model-context-protocol/apis/client-api/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/model-context-protocol/apis/client-api/
https://developers.cloudflare.com/agents/tools/codemode/mcp/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/tools/codemode/mcp/
https://developers.cloudflare.com/agents/tools/codemode/api-reference/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/tools/codemode/api-reference/
https://developers.cloudflare.com/agents/harnesses/think/configuration/	https://docs-agents-mcp-tool-materialization.preview.developers.cloudflare.com/agents/harnesses/think/configuration/

[cloudflare-docs-bot]

Review

⚠️ 1 warning, 💡 1 suggestion found in commit 180732a.

👉 Fix in your agent 👈

Fix the following review findings in PR #32180 (https://github.com/cloudflare/cloudflare-docs/pull/32180).

Before making changes, review each finding and present a brief summary table:
- For each finding, state whether you agree, disagree, or need clarification
- If you disagree (e.g. the fix requires disproportionate effort for minimal benefit,
  or the finding is factually incorrect), explain why
- If you need clarification before deciding, ask those questions
- Then share your plan for which issues to tackle and in what order

After triaging, follow this order:
1. Post a comment on this PR for any findings you are skipping, with the finding ID and your reasoning.
2. Then commit the fixes for the legitimate findings.

The comment must come before the commit — the bot reads PR comments when a new
push triggers a review, so skip comments posted after the push will be missed.

---

## Code Review

### Warnings (1)

#### CR-e34ecdaaa4b5 · Unchecked search result access
- **File:** `src/content/docs/agents/tools/codemode/durable-runtime.mdx` line 246
- **Issue:** The example does `const method = matches.results[0];` without verifying that `search()` returned any results. If no method matches, `method` is `undefined` and the next line `runtime.describe(method.path)` throws.
- **Fix:** Guard the access with a check such as `if (matches.results.length === 0) { /* handle no match */ }` before dereferencing the first result.

---

## Style Guide Review

### Suggestions (1)

#### SG-88a7fe39d425 · Use TypeScriptExample for raw TypeScript code blocks
- **File:** `src/content/docs/agents/model-context-protocol/apis/client-api.mdx` line 1155
- **Issue:** Added a raw ```ts fenced block in a docs MDX file
- **Fix:** Wrap the TypeScript snippet in <TypeScriptExample> instead of using a bare fence

Code Review

This code review is in beta and may not always be helpful — use your judgment.

Warnings (1)

File	Issue
agents/tools/codemode/durable-runtime.mdx line 246	Unchecked search result access — The example does const method = matches.results[0]; without verifying that search() returned any results. If no method matches, method is undefined and the next line runtime.describe(method.path) throws. Fix: Guard the access with a check such as if (matches.results.length === 0) { /\* handle no match \*/ } before dereferencing the first result.

Conventions

_{No convention issues found.}

Style Guide Review

Suggestions (1)

File	Issue
agents/model-context-protocol/apis/client-api.mdx line 1155	Use TypeScriptExample for raw TypeScript code blocks — Added a raw ```ts fenced block in a docs MDX file Fix: Wrap the TypeScript snippet in instead of using a bare fence

Commands

Only codeowners can run commands. Post a comment with the command to trigger it.

Command	Description
/review	Runs a review now. Incremental if a prior review exists, full if not.
/full-review	Re-reviews the entire PR diff from scratch, ignoring incremental history. Useful after a rebase, when you want a fresh review, or if the bot gets out of sync and reports issues that no longer exist.
/ignore-review-limit	Permanently lifts the 2-review automatic limit for this PR. Future pushes will trigger reviews as normal.
/disable-auto-review	Stops automatic reviews from triggering on future pushes to this PR. Codeowners can still run /review or /full-review manually.
/rebase	Rebases the PR branch against production. On conflict, attempts to resolve automatically using AI. Stops with an explanation if confidence is not high enough.