What probably matters to youSection of the release
What
A new retries setting on per-OS policyHelpers.<os> entries controls how many additional times a failing policy helper (a script that computes managed settings at startup) is retried before it's counted as a failure. A run counts as failed if the helper couldn't launch, exited with a non-zero code, or timed out. Retries use randomized backoff starting at 250ms and doubling up to 4 seconds between attempts.
Why
This gives administrators a way to tolerate transient failures, like a flaky network call inside the helper script, without immediately treating the whole policy helper as failed.
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 sinceA small documentation edit on Claude Code gateway compatibility guide touched a line naming retries after this was published.* When the upstream rejects a [thinking signature](https://platform.claude.com/docs/en/build-with-claude/extended-thinking), including with a `400` whose message says the block is `bound to a different conversation`, Claude Code removes ea…llm-gateway-protocolsee the edit
Confirmed sinceAnthropic's documentation has since written up retries, on List Events.The model is currently overloaded. Emitted after automatic retries are exhausted.api/beta/sessions/events/listsee the edit
How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agreesAnthropic's documentation has since written up retries, on List Events.
