diff --git a/.changeset/anthropic-preserved-thinking-by-default.md b/.changeset/anthropic-preserved-thinking-by-default.md deleted file mode 100644 index 1e6ad48bd..000000000 --- a/.changeset/anthropic-preserved-thinking-by-default.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Preserve prior turns' thinking by default on the Anthropic provider (Claude and Kimi's Anthropic-compatible mode), matching the Kimi default. Disable with `[thinking] keep = "off"` or `KIMI_MODEL_THINKING_KEEP=off`. diff --git a/.changeset/bg-ws-stale-recovery.md b/.changeset/bg-ws-stale-recovery.md deleted file mode 100644 index c29277144..000000000 --- a/.changeset/bg-ws-stale-recovery.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -web: Recover chat streaming after a stale background-tab WebSocket instead of requiring a page refresh. diff --git a/.changeset/fix-goal-outcome-summaries.md b/.changeset/fix-goal-outcome-summaries.md deleted file mode 100644 index 02872c198..000000000 --- a/.changeset/fix-goal-outcome-summaries.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Fix goal completion and blocked updates to produce one final user-facing outcome summary from the tool result. diff --git a/.changeset/fix-goal-queue-starts.md b/.changeset/fix-goal-queue-starts.md deleted file mode 100644 index e9f6713a1..000000000 --- a/.changeset/fix-goal-queue-starts.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Fix goal startup and queue handling so failed starts restore permission mode and queued goals wait behind new user messages. diff --git a/.changeset/fix-goal-token-budget.md b/.changeset/fix-goal-token-budget.md deleted file mode 100644 index 806efc329..000000000 --- a/.changeset/fix-goal-token-budget.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Fix goal token budgets to count model completion tokens and stop without extra continuation steps when the budget is exhausted. diff --git a/.changeset/fix-print-subagent-drain.md b/.changeset/fix-print-subagent-drain.md deleted file mode 100644 index b075e9650..000000000 --- a/.changeset/fix-print-subagent-drain.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Fix kimi -p abandoning background subagents that start late or run long, so their results reach the main agent. diff --git a/.changeset/fix-skills-dir-in-interactive-mode.md b/.changeset/fix-skills-dir-in-interactive-mode.md deleted file mode 100644 index 72b82a777..000000000 --- a/.changeset/fix-skills-dir-in-interactive-mode.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Respect the --skills-dir flag in interactive mode. diff --git a/.changeset/fix-web-btw-new-session.md b/.changeset/fix-web-btw-new-session.md deleted file mode 100644 index efe3b9d13..000000000 --- a/.changeset/fix-web-btw-new-session.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -web: Fix `/btw []` opening an empty side chat on the new-session screen. diff --git a/.changeset/fix-web-goal-new-session.md b/.changeset/fix-web-goal-new-session.md deleted file mode 100644 index a180311a9..000000000 --- a/.changeset/fix-web-goal-new-session.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -web: Fix `/goal ` silently doing nothing on the new-session screen. diff --git a/.changeset/fix-web-skill-activation-new-session.md b/.changeset/fix-web-skill-activation-new-session.md deleted file mode 100644 index 7a07e154c..000000000 --- a/.changeset/fix-web-skill-activation-new-session.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -web: Fix slash skill activations (for example `/pre-changelog`) silently doing nothing on the new-session screen. diff --git a/.changeset/hide-tool-result-system-notes.md b/.changeset/hide-tool-result-system-notes.md deleted file mode 100644 index e4a3936aa..000000000 --- a/.changeset/hide-tool-result-system-notes.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Stop showing tool-produced `` metadata in tool outputs; failed tools now show their own error text. diff --git a/.changeset/honor-explicit-anthropic-max-output.md b/.changeset/honor-explicit-anthropic-max-output.md deleted file mode 100644 index 48f84cd09..000000000 --- a/.changeset/honor-explicit-anthropic-max-output.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Honor explicit Anthropic `max_output_size` settings instead of clamping them to built-in ceilings. diff --git a/.changeset/keep-goal-tools-visible.md b/.changeset/keep-goal-tools-visible.md deleted file mode 100644 index 1a0beb393..000000000 --- a/.changeset/keep-goal-tools-visible.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Keep goal tools available to the main agent and return clear messages for invalid goal-control calls. diff --git a/.changeset/opus-4-8-max-tokens.md b/.changeset/opus-4-8-max-tokens.md deleted file mode 100644 index 3ccd00278..000000000 --- a/.changeset/opus-4-8-max-tokens.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Fix newer Claude minor versions (e.g. Opus 4.8) defaulting to the family-baseline max output tokens; an uncatalogued minor now reuses the nearest earlier known version's limit. diff --git a/.changeset/show-goal-hours.md b/.changeset/show-goal-hours.md deleted file mode 100644 index b9ebe1cf2..000000000 --- a/.changeset/show-goal-hours.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Show long-running goal wall-clock budget reminders in hours. diff --git a/.changeset/tighten-goal-mode-guidance.md b/.changeset/tighten-goal-mode-guidance.md deleted file mode 100644 index 6129ebcd0..000000000 --- a/.changeset/tighten-goal-mode-guidance.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Tighten goal-mode guidance so agents continue reasonable work across turns instead of ending goals prematurely. diff --git a/.changeset/update-permission-mode-copy.md b/.changeset/update-permission-mode-copy.md deleted file mode 100644 index 6112978f5..000000000 --- a/.changeset/update-permission-mode-copy.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Clarify the permission mode descriptions shown by `/permission`, `/auto`, and `/yolo`, and reorder `/auto` and `/yolo` in the command list. diff --git a/.changeset/wire-request-trace.md b/.changeset/wire-request-trace.md deleted file mode 100644 index dba42f11f..000000000 --- a/.changeset/wire-request-trace.md +++ /dev/null @@ -1,5 +0,0 @@ ---- -"@moonshot-ai/kimi-code": patch ---- - -Record a per-request trace in the session wire log, so model requests can be reconstructed for debugging. Not a user-facing feature. diff --git a/apps/kimi-code/CHANGELOG.md b/apps/kimi-code/CHANGELOG.md index 841cc870f..44e5829a8 100644 --- a/apps/kimi-code/CHANGELOG.md +++ b/apps/kimi-code/CHANGELOG.md @@ -1,5 +1,45 @@ # @moonshot-ai/kimi-code +## 0.23.1 + +### Patch Changes + +- [#1432](https://github.com/MoonshotAI/kimi-code/pull/1432) [`25a655c`](https://github.com/MoonshotAI/kimi-code/commit/25a655cf88b2f5861f9c0b7ea95ba9308f48d23a) Thanks [@RealKai42](https://github.com/RealKai42)! - Preserve prior turns' thinking by default on the Anthropic provider (Claude and Kimi's Anthropic-compatible mode), matching the Kimi default. Disable with `[thinking] keep = "off"` or `KIMI_MODEL_THINKING_KEEP=off`. + +- [#1451](https://github.com/MoonshotAI/kimi-code/pull/1451) [`16dc940`](https://github.com/MoonshotAI/kimi-code/commit/16dc940834d9cc693b1f0022c4c70ef0004a6102) Thanks [@wbxl2000](https://github.com/wbxl2000)! - web: Recover chat streaming after a stale background-tab WebSocket instead of requiring a page refresh. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Fix goal completion and blocked updates to produce one final user-facing outcome summary from the tool result. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Fix goal startup and queue handling so failed starts restore permission mode and queued goals wait behind new user messages. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Fix goal token budgets to count model completion tokens and stop without extra continuation steps when the budget is exhausted. + +- [#1452](https://github.com/MoonshotAI/kimi-code/pull/1452) [`244ec07`](https://github.com/MoonshotAI/kimi-code/commit/244ec077f98c2b498cee1d0002978b6963ccfd4d) Thanks [@sailist](https://github.com/sailist)! - Fix kimi -p abandoning background subagents that start late or run long, so their results reach the main agent. + +- [#1457](https://github.com/MoonshotAI/kimi-code/pull/1457) [`260a807`](https://github.com/MoonshotAI/kimi-code/commit/260a80793a95d7796950a00bdc89cf99f8b196ad) Thanks [@liruifengv](https://github.com/liruifengv)! - Respect the --skills-dir flag in interactive mode. + +- [#1445](https://github.com/MoonshotAI/kimi-code/pull/1445) [`809a88c`](https://github.com/MoonshotAI/kimi-code/commit/809a88cb34d2d5d02e43f030530bb1cd320b4a6a) Thanks [@wbxl2000](https://github.com/wbxl2000)! - web: Fix `/btw []` opening an empty side chat on the new-session screen. + +- [#1445](https://github.com/MoonshotAI/kimi-code/pull/1445) [`809a88c`](https://github.com/MoonshotAI/kimi-code/commit/809a88cb34d2d5d02e43f030530bb1cd320b4a6a) Thanks [@wbxl2000](https://github.com/wbxl2000)! - web: Fix `/goal ` silently doing nothing on the new-session screen. + +- [#1445](https://github.com/MoonshotAI/kimi-code/pull/1445) [`809a88c`](https://github.com/MoonshotAI/kimi-code/commit/809a88cb34d2d5d02e43f030530bb1cd320b4a6a) Thanks [@wbxl2000](https://github.com/wbxl2000)! - web: Fix slash skill activations (for example `/pre-changelog`) silently doing nothing on the new-session screen. + +- [#1437](https://github.com/MoonshotAI/kimi-code/pull/1437) [`743f66e`](https://github.com/MoonshotAI/kimi-code/commit/743f66e547279916d5e37454e78b11eb4b54dca3) Thanks [@RealKai42](https://github.com/RealKai42)! - Stop showing tool-produced `` metadata in tool outputs; failed tools now show their own error text. + +- [#1465](https://github.com/MoonshotAI/kimi-code/pull/1465) [`bfdbce5`](https://github.com/MoonshotAI/kimi-code/commit/bfdbce593f1dd667530cbfb5b10b5659b1968e48) Thanks [@liruifengv](https://github.com/liruifengv)! - Honor explicit Anthropic `max_output_size` settings instead of clamping them to built-in ceilings. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Keep goal tools available to the main agent and return clear messages for invalid goal-control calls. + +- [#1463](https://github.com/MoonshotAI/kimi-code/pull/1463) [`03e78ae`](https://github.com/MoonshotAI/kimi-code/commit/03e78ae19063b38119f27b1bc89097a09614c0ce) Thanks [@chengluyu](https://github.com/chengluyu)! - Fix newer Claude minor versions (e.g. Opus 4.8) defaulting to the family-baseline max output tokens; an uncatalogued minor now reuses the nearest earlier known version's limit. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Show long-running goal wall-clock budget reminders in hours. + +- [#1456](https://github.com/MoonshotAI/kimi-code/pull/1456) [`e9ef939`](https://github.com/MoonshotAI/kimi-code/commit/e9ef9399d075461c7753d1f0467bd742a81cdbf6) Thanks [@chengluyu](https://github.com/chengluyu)! - Tighten goal-mode guidance so agents continue reasonable work across turns instead of ending goals prematurely. + +- [#1450](https://github.com/MoonshotAI/kimi-code/pull/1450) [`7a65e0d`](https://github.com/MoonshotAI/kimi-code/commit/7a65e0d1c0da515dbd69f1266ba7e75713e0108e) Thanks [@liruifengv](https://github.com/liruifengv)! - Clarify the permission mode descriptions shown by `/permission`, `/auto`, and `/yolo`, and reorder `/auto` and `/yolo` in the command list. + +- [#1448](https://github.com/MoonshotAI/kimi-code/pull/1448) [`65d3017`](https://github.com/MoonshotAI/kimi-code/commit/65d30177adc11a56bdbbe9fbc3c4b92f96efd6bb) Thanks [@RealKai42](https://github.com/RealKai42)! - Record a per-request trace in the session wire log, so model requests can be reconstructed for debugging. Not a user-facing feature. + ## 0.23.0 ### Minor Changes diff --git a/apps/kimi-code/package.json b/apps/kimi-code/package.json index 68a901ab7..3d1e8916b 100644 --- a/apps/kimi-code/package.json +++ b/apps/kimi-code/package.json @@ -1,6 +1,6 @@ { "name": "@moonshot-ai/kimi-code", - "version": "0.23.0", + "version": "0.23.1", "description": "The Starting Point for Next-Gen Agents", "license": "MIT", "author": "Moonshot AI",