Follow Discord
Sweep 22 Oct 2026 · 21:04Z Build v2.1.309 528 read Stable v2.1.308 Latest v2.1.309 Next v2.1.309 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.246 ·

Credentials and storage handles passed explicitly instead of read from ambient state

Credentials and storage are passed explicitly rather than read from ambient state.

Under the hood Internal Changes
JSON All of v2.1.246
Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
InternalsArea: what it touches
Internal ChangesKind: in v2.1.246,
Internal ChangesSection of the release
What

API key verification, model-config fetching, the network-permission classifier and several resume and print helpers now take credentials and a storage handle as arguments. Worker startup and the Chrome MCP server prime credentials, cached managed settings and a stored login copy before running.

Details
  • On Windows, the credential-manager storage backend is selected only when the cached remote flag tengu_windows_credman is true, so it stays off unless that flag is set.
Evidence

api_key_verify

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

Read from
Feature flag
tengu_windows_credman Off in both readings

The flag server returned off for the account this site reads and for the anonymous baseline. A reading of off cannot rule out a rollout these two readings sit outside of.

This account: off · anonymous baseline: off · compiled default in v2.1.246: not a boolean we can read

Read once, for one account on one subscription tier, against v2.1.246. It isn't a statement about your account. What a flag value here can and cannot tell you

What has happened since
Flag reading moved The flag server now returns off for tengu_windows_credman, read as this account. A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
Flag reading moved The flag server now returns nothing at all, having stopped carrying it for tengu_windows_credman, read as this account. A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.

See this across every release →

See this entry in the whole of v2.1.246 →

Feedback