feat(codex): expose fixed gpt-5.3 reasoning aliases

feat(translator): support Claude thinking type adaptive

fix(translator): correct gemini-cli log prefix

Add Kimi (Moonshot AI) provider support

fix(test): rename test function to reflect behavior change for builtin tools

feat(thinking): enable thinking toggle for qwen3 and deepseek models

feat(registry): support provider-specific model info lookup

fix(executor): stop rewriting thinkingLevel for gemini

fix(thinking): update ValidateConfig to include fromSuffix parameter and adjust budget validation logic

refactor(thinking): add Gemini family provider grouping for strict validation

test(thinking): split E2E coverage into suffix and body parameter test functions

refactor(thinking): extract antigravity logic into a dedicated provider

feat(thinking): support zero as a valid thinking budget for capable models

test(thinking): remove legacy unit and integration tests

refactor: improve thinking logic

Merge pull request #553 from XInTheDark/fix/builtin-tools-web-search

fix(thinking): fallback to upstream model for thinking support when alias not in registry

feat(thinking): add numeric budget to thinkingLevel conversion fallback

feat(amp): add per-client upstream API key mapping support

fix: apply thinkingLevel from model suffix metadata for Gemini 3

Revert "Fix invalid thinking signature when proxying Claude via Antigravity"

Merge pull request #570 from fuguiKz/fix/antigravity-thinking-signature

Fix antigravity Claude thinking signature handling

test(thinking): add effort to budget coverage

fix(thinking): align budget effort mapping across translators

fix(translator): preserve built-in tools across openai<->responses

refactor(thinking): export thinking helpers

fix(thinking): normalize effort mapping

test(thinking): expand conversion edge case coverage

fix(thinking): centralize reasoning_effort mapping

test(thinking): cover openai-compat reasoning passthrough

fix(thinking): gate reasoning effort by model support

fix(thinking): map budgets to effort levels

refactor(api): simplify request body parsing in ampcode handlers

feat(api): add comprehensive ampcode management endpoints

refactor(config): improve OpenAI compatibility target matching logic