Moltbot

Files

Sid c9f0d6ac8e feat(agents): support thinkingDefault: "adaptive" for Anthropic models (#31227 )

* feat(agents): support `thinkingDefault: "adaptive"` for Anthropic models

Anthropic's Opus 4.6 and Sonnet 4.6 support adaptive thinking where the
model dynamically decides when and how much to think.  This is now
Anthropic's recommended mode and `budget_tokens` is deprecated on these
models.

Add "adaptive" as a valid thinking level:
- Config: `agents.defaults.thinkingDefault: "adaptive"`
- CLI: `/think adaptive` or `/think auto`
- Pi SDK mapping: "adaptive" → "medium" effort at the pi-agent-core
  layer, which the Anthropic provider translates to
  `thinking.type: "adaptive"` with `output_config.effort: "medium"`
- Provider fallbacks: OpenRouter and Google map "adaptive" to their
  respective "medium" equivalents

Closes #30880

Made-with: Cursor

* style(changelog): format changelog with oxfmt

* test(types): fix strict typing in runtime/plugin-context tests

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>

2026-03-02 03:52:02 +00:00

apply.test.ts

fix: restore dm command and self-chat auth behavior

2026-02-26 18:49:16 +01:00

apply.ts

fix: stabilize secrets land + docs note (#26155 ) (thanks @joshavant)

2026-02-26 14:47:22 +00:00

audit.test.ts

test: harden flaky timeout and resolver specs

2026-03-01 21:30:07 +00:00

audit.ts

fix: stabilize secrets land + docs note (#26155 ) (thanks @joshavant)