Create a Message changedapi/beta/messages/create
Nearest release: v2.1.315, published under an hour before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+23added
Lines−5removed
From line
135
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits33to this page, all time
The whole hunk
from line 135, old and new numbered
/
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
from line 135
135135
136136 - `"analytics-usage-cost-2026-08-24"`
137137
138 - `"experimental-claude-tag-2026-10-21"`
139
138140- `"anthropic-user-profile-id": optional string`
139141
140142 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization.
from line 1036
10341036
10351037 - `BetaEncryptedCodeExecutionResultBlockParam object`
10361038
1037 Code execution result with encrypted stdout for PFC + web_search results.
1039 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form, sent back in a later request.
10381040
10391041 - `type: "encrypted_code_execution_result"`
10401042
from line 1572
15701572
15711573 - `BetaCodeExecutionTool20260120 object`
15721574
1573 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).
1575 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.
15741576
1577 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.
1578
1579 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.
1580
15751581 - `type: "code_execution_20260120"`
15761582
15771583 - `name: "code_execution"`
from line 4224
42184224
42194225 - `BetaCodeExecutionTool20260120 object`
42204226
4221 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).
4227 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.
42224228
4229 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.
4230
4231 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.
4232
42234233 - `BetaCodeExecutionTool20260521 object`
42244234
42254235 Code execution tool with REPL state persistence.
from line 4924
49144924
49154925 - `BetaEncryptedCodeExecutionResultBlock object`
49164926
4917 Code execution result with encrypted stdout for PFC + web_search results.
4927 The output and exit code of Python code that Claude ran with the code execution tool, with standard output in encrypted form.
49184928
4929 The API returns this block in place of `code_execution_result` when the code called the web search tool.
4930
4931 See the [programmatic tool calling guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling?sdk_languages=http) to learn more about programmatic tool calling.
4932
49194933 - `type: "encrypted_code_execution_result"`
49204934
49214935 default: encrypted_code_execution_result
from line 5490
54765490
54775491 - `BetaCodeExecutionTool20260120 object`
54785492
5479 Code execution tool with REPL state persistence (daemon mode + gVisor checkpoint).
5493 A `tools` entry for the code execution tool, version `code_execution_20260120`. It adds REPL state persistence to version `code_execution_20250825`.
54805494
5495 REPL state persistence applies with programmatic tool calling. With it, the Python interpreter state, such as variable bindings, persists across requests that reuse the container.
5496
5497 See the [code execution tool guide](https://platform.claude.com/docs/en/agents-and-tools/tool-use/code-execution-tool?sdk_languages=http#tool-versions) to learn what each tool version adds.
5498
54815499 - `type: "code_execution_20260120"`
54825500
54835501 - `name: "code_execution"`
from line 7018
70007018
70017019 - `response_inclusion: optional "full" or "excluded"`
70027020
7003 How this tool's result blocks appear in the API response when the result was consumed by a completed code_execution call in the same turn. 'full' returns the complete content (default). 'excluded' drops the nested server_tool_use and result block pair entirely. Results from direct calls, or from code_execution calls that paused before completing, are always returned in full so they can be sent back on the next turn.
7004
7005 - `"full"`
7006
7007 - `"excluded"`
7008
7009 - `strict: optional boolean`
7010
7011 When true, guarantees schema validation on tool names and inputs
7012
7013 - `url_sources: optional BetaWebFetchURLSources or null`
7014
7015 Which sources contribute to the set of URLs the tool may fetch. Omitted means every source.
7016
7017 - `use_cache: optional boolean`
7018
7019 Whether to use cached content. Set to false to bypass the cache and fetch fresh content. Only set to false when the user explicitly requests fresh content or when fetching rapidly-changing sources.
7020
7021 - `BetaAdvisorTool20260301 object`
7022
7023 - `type: "advisor_20260301"`
7024
7025 - `model: Model`
7026
7027 The model that will complete your prompt.
7028
7029 See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options.
7030
7031 - `"claude-haiku-5-5"`
7032
7033 Fastest model for high-volume, real-time tasks
7034
7035 - `"claude-sonnet-5-5"`
7036
7037 Efficient model for coding and agents
7038
7039 - `"claude-fable-5-1"`
7040
7041 Frontier intelligence for ambitious tasks across coding, scientific discovery, and enterprise workflows
7042
7043 - `"claude-opus-5-5"`
7044
7045 Powerful intelligence for coding, knowledge work, and long-running agents
7046
7047
7021 How this tool's result blocks appear in the
No line in this hunk matches that.