[DOCS] MCP/WebFetch docs missing binary output file-handling behavior introduced in v2.1.69

Documentation Type

Missing documentation (feature not documented)

Documentation Location

https://code.claude.com/docs/en/mcp

Section/Topic

Binary tool output handling (MCP tools and WebFetch).

Current Documentation

Tool reference docs currently provide only high-level tool descriptions, for example:

WebFetch | Fetches content from a specified URL

No documentation currently describes binary-response handling (decoded file output, file extensions, or where files are written).

What's Wrong or Missing?

Changelog v2.1.69 introduced a user-visible behavior change:

• Binary MCP/WebFetch responses (PDF, Office docs, audio) are saved as decoded files with correct extensions
• Raw base64 is no longer dumped into conversation context

This behavior is not documented in MCP/tool docs.

Suggested Improvement

Add a "Binary output handling" section covering:

• Which content types are written to files
• Where files are saved
• How filenames/extensions are chosen
• How this appears in CLI output and transcripts

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Context |
|------|---------|
| https://code.claude.com/docs/en/mcp | Primary MCP behavior docs should describe binary-output semantics |
| https://code.claude.com/docs/en/settings | Tool reference includes WebFetch but not binary file output behavior |
| https://platform.claude.com/docs/en/agent-sdk/overview | Agent SDK cross-reference for tool output semantics |

Total scope: 3 pages affected

Source: Changelog v2.1.69

Exact changelog entry:

Improved MCP binary content handling: tools returning PDFs, Office documents, or audio now save decoded bytes to disk with the correct file extension instead of dumping raw base64 into the conversation context. WebFetch also saves binary responses alongside its summary.