Commit graph

127 commits

Author SHA1 Message Date
github-actions[bot]
696b46ed11
ci: release packages (#453)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-06-05 18:25:21 +08:00
liruifengv
f61165aa9a
fix: use underscore sub_skill experimental flag id (#483)
* fix: use underscore sub_skill experimental flag id

* chore: keep changeset wording unchanged
2026-06-05 18:01:34 +08:00
_Kerman
420a387e21
chore: remove llm request token estimate (#479) 2026-06-05 17:19:46 +08:00
Haozhe
df4f2d6e86
feat(skills): add experimental sub-skill discovery (#468)
- Add sub-skill feature flag and recursive bundle discovery behind KIMI_CODE_EXPERIMENTAL_SUB_SKILL.
- Register builtin sub-skill review and consolidate helpers as inline hidden skills.
2026-06-05 16:10:41 +08:00
_Kerman
aa610e247d
feat: use fixed 30-minute subagent timeout (#470) 2026-06-05 15:59:10 +08:00
Kai
93eb70a727
feat(env): migrate kimi-cli model request params and auto-update toggle (#458)
* feat(env): migrate kimi-cli model request params and auto-update toggle

Migrate still-relevant environment variables from kimi-cli:

- KIMI_MODEL_TEMPERATURE / KIMI_MODEL_TOP_P: sampling params applied
  globally to any kimi provider (not tied to KIMI_MODEL_NAME).
- KIMI_MODEL_THINKING_KEEP: Moonshot preserved-thinking passthrough
  (thinking.keep), injected only while Thinking is on.
- KIMI_CODE_NO_AUTO_UPDATE (legacy alias KIMI_CLI_NO_AUTO_UPDATE):
  fully disables the update preflight.

Wires env -> provider in Agent.get llm() via applyKimiEnvGenerationParams,
reusing kosong's existing GenerationKwargs / thinking.keep support.
KIMI_MODEL_MAX_TOKENS is intentionally untouched: it already flows through
the completion-budget path.

* fix(env): apply Kimi sampling params to compaction requests too

Sink KIMI_MODEL_TEMPERATURE / KIMI_MODEL_TOP_P into ConfigState.provider so
every request built from config.provider — main loop and full-history
compaction alike — carries them, matching kimi-cli where these live on the
shared create_llm provider. thinking.keep stays in Agent.llm because it
depends on the runtime thinking state (compaction runs thinking-off and
correctly skips it).

Splits applyKimiEnvGenerationParams into applyKimiEnvSamplingParams (applied
at provider construction) and applyKimiEnvThinkingKeep (applied in Agent.llm).

Addresses PR review feedback about compaction requests bypassing the wrapped
provider.
2026-06-05 14:54:24 +08:00
_Kerman
95dd0a1f3b
chore: clarify agent record resume contract (#465) 2026-06-05 14:51:58 +08:00
_Kerman
4f9977d4dc
fix: preserve thinking during compaction (#464) 2026-06-05 14:25:32 +08:00
7Sageer
1fe5d5549c
fix(kosong): clamp OpenAI chat xhigh effort by model (#457) 2026-06-05 14:01:50 +08:00
_Kerman
3a98713050
fix: show concise filtered response errors (#456) 2026-06-05 12:48:35 +08:00
github-actions[bot]
12d062d48e
ci: release packages (#390)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-06-04 21:45:01 +08:00
liruifengv
85338e9f7d
feat: add built-in update-config skill (#433)
Inline, model-invocable skill to inspect and safely edit kimi-code's own config.toml and tui.toml (copy to a candidate, Edit, validate with kimi doctor, timestamped backup, then overwrite). Also documents tui.toml on the config files page (en + zh).
2026-06-04 21:41:54 +08:00
liruifengv
6a4e4c75d4
feat(cli): add doctor command for config validation (#431)
* feat(cli): add doctor command for config validation

* fix(cli): format doctor validation errors

* fix(cli): validate doctor config through SDK RPC

* chore(changeset): simplify doctor release note
2026-06-04 21:15:18 +08:00
Luyu Cheng
beb12ac021
feat: add upcoming goal queue (#393) 2026-06-04 20:43:04 +08:00
7Sageer
7856c03176
fix: avoid runtime check rpc (#432) 2026-06-04 20:37:17 +08:00
7Sageer
be0da5ff39
fix: fail early when Git Bash is missing (#430)
* docs: mention Windows Git Bash requirement

* docs: add Windows Git Bash changeset

* fix: fail early when Git Bash is missing

* test: update goal prompt harness mock
2026-06-04 20:16:12 +08:00
liruifengv
86a42a26a1
feat: Add persistent experimental feature toggles and a TUI panel (#420)
* feat: Add persistent experimental feature toggles and a TUI panel

* test(agent-core): isolate experimental flag tests

* fix(tui): block experiments panel while busy

* refactor(experimental): merge feature query APIs
2026-06-04 18:35:41 +08:00
7Sageer
459826292f
fix: normalize malformed responses rate limit errors (#411) 2026-06-04 16:21:25 +08:00
liruifengv
15d71b5130
feat: add reload commands (#383)
* Create CONFIG_RELOAD_RESEARCH.md

* feat: add reload commands for session and tui

* Delete CONFIG_RELOAD_RESEARCH.md

* fix(tui): refresh terminal title on reload

---------

Co-authored-by: qer <wbxl2000@outlook.com>
2026-06-04 13:29:12 +08:00
7Sageer
232ed874d4
fix: scope managed OAuth credentials (#399)
* fix: scope managed OAuth credentials

* refactor: centralize managed OAuth auth resolution
2026-06-04 13:14:18 +08:00
Nekof
6e74027fdc
fix: lowercase edit stale-content message (#387) 2026-06-03 23:20:27 +08:00
github-actions[bot]
6c0afc4d9c
ci: release packages (#360)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-06-03 22:00:17 +08:00
7Sageer
76441c21fb
fix: load root mcp json config (#382) 2026-06-03 21:13:46 +08:00
Haozhe
3eafa79f39
feat(acp): implement ACP server with session lifecycle, tool streaming, and IDE integration (#368)
This commit scaffolds the @moonshot-ai/acp-adapter package and introduces
the full ACP (Agent Communication Protocol) server implementation for
Kimi Code CLI, including:

- Scaffold @moonshot-ai/acp-adapter workspace package with build skeleton
- `kimi acp` CLI subcommand and stdout-safe logging
- ACP version negotiation and AgentSideConnection wrapper
- Auth gate for session creation
- Session lifecycle: new, list, load with history replay
- Prompt content conversion (text, image, embedded resources, resource links)
- Assistant streaming with thinking support and end-turn handling
- Tool call streaming (started, delta, progress) with result conversion (text / diff)
- Approval handling with diff/text display blocks mapped to ACP options
- Kaos read/write interface (AcpKaos) for unsaved buffer access
- Session mode (yolo/auto) and model management
- Config options builder with thinking toggle
- MCP server forwarding from ACP to harness
- Agent plan updates and available commands updates
- AskUserQuestion bridged to session/request_permission
- Plan review options surfaced through requestPermission
- Error mapping, ext_method stubs, and graceful shutdown
- IDE integration guide (Zed + JetBrains)
- End-to-end tests against ACP TS SDK client
2026-06-03 21:11:30 +08:00
_Kerman
8639105313
fix: resume subagents lazily (#380) 2026-06-03 20:16:43 +08:00
Luyu Cheng
6a2252343a
fix: simplify goal budget schema and fix output caps (#365) 2026-06-03 15:55:14 +08:00
_Kerman
ba7dd736a3
feat: add btw side-channel command (#338) 2026-06-03 15:54:31 +08:00
_Kerman
a6b16ce6b4
refactor: split SDK RPC client (#339) 2026-06-03 15:30:38 +08:00
Haozhe
9e6972423e
feat(agent-core): detect and escalate repeated tool calls to prevent infinite loops (#364)
* feat(agent-core): add dead-end detection for repeated tool calls

- inject stronger dead-end reminder at streak 10 and force-stop turn at streak 15
- emit `tool_call_repeat` telemetry events for streak >= 2
- pass telemetry client from TurnFlow to ToolCallDeduplicator
- add comprehensive tests for dead-end stop and telemetry behavior

* fix(agent-core): tune tool-dedup thresholds and preserve isError state

- Lower repeat-reminder thresholds to 3, 5, 8 and force-stop to 12
- Force-stop no longer overrides successful tool results to isError
- Move stopTurn field into ExecutableToolSuccessResult for explicit loop-control without requiring isError
2026-06-03 14:16:14 +08:00
_Kerman
33496bd031
test: disable harness log writes by default (#362) 2026-06-03 12:55:54 +08:00
_Kerman
179aecf423
feat: log enabled experimental flags (#357) 2026-06-03 11:35:29 +08:00
github-actions[bot]
14cd8d098b
ci: release packages (#296)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-06-02 22:54:57 +08:00
Luyu Cheng
ac37d74484
feat: pursue a goal autonomously (#270) 2026-06-02 22:52:23 +08:00
_Kerman
191059d400
feat: add background ask user questions (#315) 2026-06-02 22:12:25 +08:00
qer
3502d870fc
chore: scrub internal identifiers from public text (#340) 2026-06-02 21:54:42 +08:00
Haozhe
7284f30479
fix(oauth): fix multi-provider custom registry import losing providers (#335)
- Add `applyCustomRegistryEntries` helper to apply all entries in-memory before a single write
- Update CLI Add Platform flow to use it, replacing interleaved in-memory mutations with removeProvider RPCs
- Add regression tests for repeated imports and provider field refreshing
2026-06-02 20:05:00 +08:00
qer
7cda9c3866
feat: add permission approval hooks (#336) 2026-06-02 19:57:10 +08:00
Haozhe
80164c2e97
fix(glob): normalize pattern before brace expansion (#311)
* fix(glob): normalize pattern before brace expansion

* fix(glob): preserve backslash-escaped metacharacters and normalize braces correctly

- expand braces before normalizing to prevent `..` from collapsing across brace groups
- skip normalization for patterns containing backslash-escaped glob metacharacters
- add test coverage for escaped braces and nested `..` inside alternatives
2026-06-02 19:30:32 +08:00
7Sageer
7ffb5dd9b3
fix: drop unsigned thinking for Claude replay (#331) 2026-06-02 18:20:10 +08:00
liruifengv
1178c5cd14
feat: add todo list reminder injection (#333) 2026-06-02 18:12:38 +08:00
_Kerman
7a47045af2
fix: inherit custom tools in subagents (#330) 2026-06-02 17:16:59 +08:00
7Sageer
8809f3eb11
fix: normalize tool call ids across providers (#327) 2026-06-02 16:42:40 +08:00
_Kerman
fe7db4a7e3
feat: append TODO list as markdown to compaction summary (#319) 2026-06-02 16:20:52 +08:00
Haozhe
0d074f1bae
feat(agent-core): propagate app version into agent record metadata (#308)
* feat(agent-core): propagate app version into agent record metadata

- add appVersion option to AgentOptions, SessionOptions, and KimiCoreOptions
- forward appVersion through Agent, Session, and SDKRpcClient
- include app_version in agent record metadata events
2026-06-02 14:54:29 +08:00
_Kerman
a217ff09aa
feat: add /undo slash command and keep replay in sync (#277) 2026-06-02 12:48:59 +08:00
_Kerman
573c56e829
refactor: background task manager persistence (#285) 2026-06-02 12:48:31 +08:00
_Kerman
91b292e898
fix: allow glob outside workspace (#283) 2026-06-02 12:36:16 +08:00
opbr
d912053b0d
fix(kaos): search usr\bin\bash.exe paths for Git Bash on Windows (#145)
Some Git for Windows installations have bash.exe only at
<root>\usr\bin\bash.exe while <root>\bin\bash.exe does not exist.
Add usr\bin\bash.exe candidates alongside each bin\bash.exe entry in
both the hardcoded fallback list and the git.exe inference logic.

Also add @moonshot-ai/kimi-code-sdk to the changeset since the SDK
bundles agent-core code and calls detectEnvironmentFromNode() on
session creation, so Windows SDK users also need this fix.
2026-06-02 11:41:19 +08:00
zha-ji-tui
0071b63fc8
fix(skill): wrap slash-activated skill content in system-reminder tags (#135) 2026-06-02 11:20:54 +08:00
github-actions[bot]
121a6dd2a1
ci: release packages (#237)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-06-02 10:22:31 +08:00