What
Settings expose three policies for AWS requests passing through the proxy, covering streaming, presigned and sigv4a requests. Each accepts deny or passthrough, and deny is documented as the default: the request fails closed with a 403. passthrough forwards it without re-signing, which the upstream service will reject on signature grounds.
Evidence
` deny (default) fails closed with a 403; passthrough forwards the request unre-signed (the upstream will reject its signature). `
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
Names in the bundledenypassthrough
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Added since
A small documentation edit on Voice dictation touched a line naming passthrough after this was published.
* **`Voice input is failing repeatedly and has been paused`**: voice dictation hit three failures within 10 seconds. Claude Code pauses dictation until 10 seconds have passed since the first of those failures. This usually means the microp…voice-dictation see the edit
Confirmed since
Anthropic's documentation has since written up passthrough, on User Profiles.
How the platform uses the API for this entity. `application` (default): the profile represents an individual end-user of the platform's product. `passthrough`: the profile identifies a company the platform resells Claude access to.api/beta/user_profiles see the edit
Added since
A small documentation edit on multiagent-orchestration touched a line naming deny after this was published.
Under `auto`, your `user.message` events can lead the server to allow a call it would otherwise deny. Nothing in a subagent's thread counts as your intent: your client posts no messages there, and the coordinator's messages to the subagent…managed-agents/multiagent-orchestration see the edit
Confirmed since
Anthropic's documentation has since written up deny, on Archive Session.
The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-…api/beta/sessions/archive see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
Anthropic's documentation has since written up deny, on Archive Session.