jeo-code 0.6.21 → 0.6.22

package/CHANGELOG.md

@@ -5,7 +5,16 @@ The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 The README mirrors the latest 5 entries — regenerate with `bun run changelog:sync`.
-+
+
+## [0.6.22] - 2026-06-18
+_Extended-thinking activation is now consistent across providers: a `low` session thinking level enables reasoning everywhere._
+
+### Changed
+- **Anthropic now enables extended thinking at `low` effort, matching Gemini and OpenAI.** Previously Anthropic disabled extended thinking for `low`/`minimal`/unset effort while Gemini (`thinkingConfig.thinkingBudget`) and OpenAI (`reasoning_effort`) kept reasoning on at `low`, so the same session thinking level produced thinking on some providers but not Anthropic. `anthropicThinkingBudget` now maps `low → 4000` budget_tokens (same tier as Gemini), with `medium → 10000` and `high → 24000` unchanged; only `minimal`/unset stay non-thinking so `/fast` and minimal thinking remain cheaper/faster across all providers.
+
+### Added
+- **Anthropic `low`-effort thinking-parity test** (`test/anthropic-stream.test.ts`) asserting `anthropicPayload` emits `thinking.budget_tokens` for low/medium/high and omits it for minimal/unset.
+
 ## [0.6.21] - 2026-06-18
 _Session thinking level now reaches the provider's actual reasoning depth, not just the token ceiling._
 

package/README.ja.md

@@ -200,11 +200,11 @@ CI は `.github/workflows/npm-publish.yml` で公開します — GitHub リリ
 ## 変更履歴 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.6.22]** (2026-06-18) — Extended-thinking activation is now consistent across providers: a `low` session thinking level enables reasoning everywhere.
 - **[0.6.21]** (2026-06-18) — Session thinking level now reaches the provider's actual reasoning depth, not just the token ceiling.
 - **[0.6.20]** (2026-06-18) — Launch REPL internals decomposed into testable modules: `@mention` path completion, slash-command view renderers, and slash-command handlers extracted from the monolithic `launch.ts` into dedicated files with full unit-test coverage.
 - **[0.6.19]** (2026-06-18) — Post-turn hooks run once per batch (not per edit), local hook reads are mtime-cached, tool-result formatting is parallelized, and wrapped colored text keeps its tint.
 - **[0.6.18]** (2026-06-17) — Memory data-flow diagram and a README "Memory flow" section documenting the actual runtime behavior.
-- **[0.6.17]** (2026-06-17) — Legacy MEMORY.md migrates losslessly into the OKF concept bundle, with a one-shot command and a rollback toggle.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.ko.md

@@ -200,11 +200,11 @@ CI는 `.github/workflows/npm-publish.yml`로 배포합니다 — GitHub 릴리
 ## 변경 이력 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.6.22]** (2026-06-18) — Extended-thinking activation is now consistent across providers: a `low` session thinking level enables reasoning everywhere.
 - **[0.6.21]** (2026-06-18) — Session thinking level now reaches the provider's actual reasoning depth, not just the token ceiling.
 - **[0.6.20]** (2026-06-18) — Launch REPL internals decomposed into testable modules: `@mention` path completion, slash-command view renderers, and slash-command handlers extracted from the monolithic `launch.ts` into dedicated files with full unit-test coverage.
 - **[0.6.19]** (2026-06-18) — Post-turn hooks run once per batch (not per edit), local hook reads are mtime-cached, tool-result formatting is parallelized, and wrapped colored text keeps its tint.
 - **[0.6.18]** (2026-06-17) — Memory data-flow diagram and a README "Memory flow" section documenting the actual runtime behavior.
-- **[0.6.17]** (2026-06-17) — Legacy MEMORY.md migrates losslessly into the OKF concept bundle, with a one-shot command and a rollback toggle.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.md

@@ -200,11 +200,11 @@ Required npm token permissions (repository secret `NPM_TOKEN`):
 ## Changelog
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.6.22]** (2026-06-18) — Extended-thinking activation is now consistent across providers: a `low` session thinking level enables reasoning everywhere.
 - **[0.6.21]** (2026-06-18) — Session thinking level now reaches the provider's actual reasoning depth, not just the token ceiling.
 - **[0.6.20]** (2026-06-18) — Launch REPL internals decomposed into testable modules: `@mention` path completion, slash-command view renderers, and slash-command handlers extracted from the monolithic `launch.ts` into dedicated files with full unit-test coverage.
 - **[0.6.19]** (2026-06-18) — Post-turn hooks run once per batch (not per edit), local hook reads are mtime-cached, tool-result formatting is parallelized, and wrapped colored text keeps its tint.
 - **[0.6.18]** (2026-06-17) — Memory data-flow diagram and a README "Memory flow" section documenting the actual runtime behavior.
-- **[0.6.17]** (2026-06-17) — Legacy MEMORY.md migrates losslessly into the OKF concept bundle, with a one-shot command and a rollback toggle.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/README.zh.md

@@ -200,11 +200,11 @@ CI 通过 `.github/workflows/npm-publish.yml` 发布 — GitHub 发布 release
 ## 更新日志 (Changelog)
 
 <!-- CHANGELOG:START (auto-generated from CHANGELOG.md — run `bun run changelog:sync`) -->
+- **[0.6.22]** (2026-06-18) — Extended-thinking activation is now consistent across providers: a `low` session thinking level enables reasoning everywhere.
 - **[0.6.21]** (2026-06-18) — Session thinking level now reaches the provider's actual reasoning depth, not just the token ceiling.
 - **[0.6.20]** (2026-06-18) — Launch REPL internals decomposed into testable modules: `@mention` path completion, slash-command view renderers, and slash-command handlers extracted from the monolithic `launch.ts` into dedicated files with full unit-test coverage.
 - **[0.6.19]** (2026-06-18) — Post-turn hooks run once per batch (not per edit), local hook reads are mtime-cached, tool-result formatting is parallelized, and wrapped colored text keeps its tint.
 - **[0.6.18]** (2026-06-17) — Memory data-flow diagram and a README "Memory flow" section documenting the actual runtime behavior.
-- **[0.6.17]** (2026-06-17) — Legacy MEMORY.md migrates losslessly into the OKF concept bundle, with a one-shot command and a rollback toggle.
 
 See [CHANGELOG.md](CHANGELOG.md) for the full history.
 <!-- CHANGELOG:END -->

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "jeo-code",
-  "version": "0.6.21",
+  "version": "0.6.22",
   "description": "Clean, highly optimized AI coding agent using spec-first loop",
   "type": "module",
   "main": "src/cli.ts",

package/src/ai/providers/anthropic.ts

@@ -72,11 +72,13 @@ function anthropicSystemBlocks(
   return blocks;
 }
 
-/** Anthropic extended-thinking budget by reasoning effort (kept under max_tokens). Off for
- *  low/minimal/unset effort so /fast and minimal thinking stay non-thinking (cheaper/faster). */
+/** Anthropic extended-thinking budget by reasoning effort (kept under max_tokens). Cross-provider
+ *  parity (matches Gemini's tiers): low/medium/high all enable thinking with scaling depth; only
+ *  minimal/unset stay non-thinking so /fast and minimal thinking remain cheaper/faster. */
 function anthropicThinkingBudget(effort: CallOptions["reasoningEffort"], maxTokens: number): number | undefined {
   let budget: number;
   switch (effort) {
+    case "low": budget = 4000; break;
     case "medium": budget = 10000; break;
     case "high": budget = 24000; break;
     default: return undefined;