Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All of v2.1.97 Home All releases olderv2.1.96 v2.1.98newer
Claude Code v2.1.97

Effort Level Clamping [In Development]

What

Infrastructure to cap the maximum effort level based on a model's capability tier, preventing users from setting effort higher than what a model supports.

Status

Feature-flagged (disabled by default)

Details
  • Gated behind tengu_pyrite_wren flag with default !1 (false)
  • Compares the requested effort against the model's position in the effort ranking
Evidence

Function Uz4 gated by R8("tengu_pyrite_wren", !1) (search for "tengu_pyrite_wren")

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

See this entry in the whole of v2.1.97 →