mirror of
https://github.com/MoonshotAI/kimi-code.git
synced 2026-07-09 17:29:12 +00:00
65d30177ad
* feat(agent-core): record llm request trace in wire.jsonl Add three observability record types so every request sent to the model can be reconstructed from the wire log at the logical-request level: - llm.tools_snapshot: content-addressed snapshot of the top-level tools table as sent (post deferred-strip), written once per unique table - llm.request: one record per outbound request (retries, strict resends, and compaction rounds included) carrying the effective request params and hash links to the system prompt and tools snapshot - mcp.tools_discovered: the server's verbatim tools/list result plus the agent's gating (allow-list, collisions), deduplicated by content hash Observability records never feed state rebuild; replay only restores the write-dedup cursors. The records/types.ts contract now documents the two record classes explicitly (persisted is not the same as replayed). Recording happens at the single Agent.generate choke point. The LLMRequestLogFields side channel gains kind/projection/maxTokens/ droppedCount, chatWithRetry preserves caller-set fields, and compaction tags its requests. The vis wire view renders the new record kinds. * fix(agent-core): record the provider-clamped completion cap in the request trace The llm.request trace recorded the client-requested budget cap, but chat-completions providers tighten the actual wire value inside withMaxCompletionTokens (remaining-context sizing, transport ceilings, model-default resolution) — with the default budget the clamp is active on nearly every non-empty-context request, so the recorded value did not match what was sent. Providers now expose the effective cap they computed as a readonly maxCompletionTokens field on the clone, and the recorder reads it from the effective provider at the Agent.generate choke point. This replaces the side-channel recomputation, which is removed along with the appliedCompletionBudgetCap helper. * fix(agent-core): park pre-replay MCP discovery records and hash the collision outcome Two wire-hygiene fixes for the mcp.tools_discovered trace: Parking: the real Session ordering connects MCP servers concurrently with agent construction, so ToolManager can observe a connected server before agent.resume() has replayed the wire. Recording at that point bypassed the restored dedup cursor (duplicating a 1-50KB record on every resume) and appended a stray metadata record ahead of replay. AgentRecords now exposes a one-shot opened latch — set when replay completes (after the migration rewrite flushes) or when the first live record is logged — and ToolManager parks discoveries until then, re-running the dedup check at drain time. A frozen range-limited replay never opens; those agents are transient previews. Collision hashing: the dedup hash now covers the collision outcome, not just the raw list and allow-list. Collisions depend on which other servers hold a sanitized qualified name at registration time, so a server can re-register with identical tools but a flipped outcome; that gating change must produce a new record instead of being suppressed. * fix(agent-core): skip the request trace for pre-flight-aborted calls Mirror kosong generate()'s pre-flight abort check at the Agent.generate choke point: a call whose signal is already aborted never reaches the wire (generate throws before dispatching), so it must not leave an llm.request/llm.tools_snapshot trace or a diagnostic log line claiming a request was sent. Recording stays before dispatch for every call that passes the gate, preserving the crash-safety of the trace. * chore(agent-core): remove a leftover adaptive-thinking override hook The adaptiveThinkingOverride option was a temporary local hook explicitly marked for removal before commit. Nothing passes it, so resolution falls back to the alias-level adaptiveThinking value in all cases; drop the option and the dead indirection. * fix(kosong): derive the exposed completion cap from generation kwargs maxCompletionTokens was a field stored only by withMaxCompletionTokens, so caps that reach the wire through other paths were invisible to the request trace: with completion budgeting disabled via env, Anthropic still sends the constructor-resolved max_tokens (required by the Messages API), and constructor-level kwargs like OpenAILegacyOptions maxTokens were likewise unreported. Replace the stored field with a getter derived from each provider's generation kwargs — the single source the request body reads — covering constructor defaults, direct withGenerationKwargs configuration, and budget application in one place. Kimi mirrors its request-time legacy max_tokens alias normalization; openai-legacy reuses the same normalizeGenerationKwargs the request path uses. * feat(agent-core): add thinkingKeep passthrough for Kimi providers and update tests
3.8 KiB
3.8 KiB
会话与上下文
Kimi Code CLI 把每次对话持久化为一个「会话」，保留消息历史和元数据，可以随时关闭终端后再回来继续。本页介绍如何恢复会话、管理上下文，以及导出和派生会话。
会话存储
所有会话保存在 $KIMI_CODE_HOME/sessions/ 下（默认 ~/.kimi-code/sessions/），按工作目录分组存放：
~/.kimi-code/
├── config.toml
├── session_index.jsonl
└── sessions/
└── <workDirKey>/
└── <sessionId>/
├── state.json
└── agents/
├── main/
│ └── wire.jsonl
└── <subagentId>/
└── wire.jsonl
state.json：会话标题、创建时间等元数据。agents/*/wire.jsonl：Agent 事件流，用于会话恢复和回放；同时记录发给模型的请求轨迹（工具 schema、请求参数、MCP 工具清单），便于调试。
::: warning 注意
sessions/ 目录下的文件请勿手动编辑，否则可能导致会话无法正常恢复。
:::
启动与恢复会话
每次直接运行 kimi 都会创建新会话。以下方式可以恢复历史会话：
继续当前目录最近的会话：
kimi --continue
恢复指定会话（通过 ID）：
kimi --session abc123
交互式浏览历史会话并选择：
kimi --session
::: warning 注意
--continue 与 --session 互斥。
:::
在 TUI 中切换会话
不离开当前终端也可以管理会话，以下斜杠命令仅在 Agent 空闲时可用：
/new（别名/clear）：切换到新会话，丢弃当前上下文。/sessions（别名/resume）：浏览并恢复历史会话。/fork：派生当前会话（详见下文）。/title <text>（别名/rename）：设置会话标题方便识别；不带参数时显示当前标题。
上下文压缩
对话变长时，Kimi Code CLI 会在上下文接近窗口上限时自动压缩历史消息，释放 token 空间。也可以随时手动触发：
/compact
压缩时可以附带指引，告诉模型优先保留哪些信息：
/compact 保留与数据库迁移相关的讨论
派生会话
想在不破坏当前对话的前提下尝试新思路，使用 /fork：
/fork
派生后的两个会话彼此独立，互不影响，可以随时通过 /sessions 切回原来的会话。已保存的 /goal 不会复制到派生会话。如果你想在派生会话中进行自主 goal 工作，需要在那里开始一个新 goal。
导出会话
用 kimi export 把会话打包为 ZIP，适合分享、归档或提交问题反馈：
kimi export <sessionId>
不传 sessionId 时导出当前目录最近的会话（有交互式确认，加 -y 跳过）。用 -o 指定输出路径：
kimi export <sessionId> -o ~/Desktop/my-session.zip
导出包含会话目录下的所有文件，包括诊断日志。全局诊断日志（~/.kimi-code/logs/kimi-code.log）默认也会打包；如不需要，加 --no-include-global-log 排除。
也可以在 TUI 内导出，无需离开交互界面：
/export-debug-zip：产生与kimi export相同的调试 ZIP。/export-md（别名/export）：导出为人类可读的 Markdown 对话记录，适合分享或存档。可选接收路径参数；不带参数时写入工作目录下的kimi-export-<short-id>-<timestamp>.md。
::: tip 提示 导出文件可能包含代码、命令输出和路径等敏感信息，分享前请先确认内容。 :::