feat(agent-core-v2): add tokenCounting service with strategy config and measured anchors

.agents/skills/agent-core-dev/edge-exposure.md

@@ -105,7 +105,7 @@ Read = `GET`, write = `POST`. `sid` = `session_id`, `aid` = `agent_id`.
 | `tasks` | `list` / `get` / `readOutput` | IBackgroundService.* | GET |
 | `tasks` | `stop` / `detach` | IBackgroundService.* | POST |
 | `usage` | `status` | IUsageService.status | GET |
-| `context` | `status` | IAgentContextSizeService.get | GET |
+| `context` | `status` | IAgentTokenCountingService.get | GET |
 | `swarm` | `isActive` | ISwarmService.isActive | GET |
 | `swarm` | `enter` / `exit` | ISwarmService.* | POST |
 | `permission` | `getMode` | IPermissionModeService.mode | GET |

.changeset/tidy-llamas-attend.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Fix the context window limit showing as 0 in session status updates when no model is bound yet or the configured model no longer resolves; the limit now falls back to the default model or is omitted when unknown.

.changeset/token-counting-reporting-edge.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": minor
+---
+
+The `[token_counting]` strategy now only selects the reported context size: `estimated` keeps provider-reported usage out of the context-size display, and `measured` no longer gets stuck retrying an oversized compaction request until it fails.

.changeset/token-counting-strategy.md

@@ -0,0 +1,5 @@
+---
+"@moonshot-ai/kimi-code": patch
+---
+
+Add a `[token_counting]` config section to choose how context token counts are derived: `measured+estimated` (default), `measured` (provider usage only), or `estimated` (heuristic only, for providers without usage reporting). Set `strategy` under `[token_counting]` in config.toml (or `KIMI_TOKEN_COUNTING_STRATEGY`) to switch.

apps/kimi-inspect/src/panels.ts

@@ -17,7 +17,6 @@
  */
 
 import { IAgentActivityView } from '@moonshot-ai/agent-core-v2/agent/activityView/activityView';
-import { IAgentContextSizeService } from '@moonshot-ai/agent-core-v2/agent/contextSize/contextSize';
 import { IAgentGoalService } from '@moonshot-ai/agent-core-v2/agent/goal/goal';
 import { IAgentMcpService } from '@moonshot-ai/agent-core-v2/agent/mcp/mcp';
 import { IAgentPermissionModeService } from '@moonshot-ai/agent-core-v2/agent/permissionMode/permissionMode';
@@ -27,6 +26,7 @@ import { IAgentProfileService } from '@moonshot-ai/agent-core-v2/agent/profile/p
 import { IAgentRPCService } from '@moonshot-ai/agent-core-v2/agent/rpc/rpc';
 import { IAgentSwarmService } from '@moonshot-ai/agent-core-v2/agent/swarm/swarm';
 import { IAgentTaskService } from '@moonshot-ai/agent-core-v2/agent/task/task';
+import { IAgentTokenCountingService } from '@moonshot-ai/agent-core-v2/agent/tokenCounting/tokenCounting';
 import { IAgentToolRegistryService } from '@moonshot-ai/agent-core-v2/agent/toolRegistry/toolRegistry';
 import { IAgentUsageService } from '@moonshot-ai/agent-core-v2/agent/usage/usage';
 import { IAuthSummaryService } from '@moonshot-ai/agent-core-v2/app/auth/auth';
@@ -179,8 +179,8 @@ export const AGENT_PANELS: readonly ServicePanelDef[] = [
     fetch: (svc) => call(svc, 'status'),
   },
   {
-    id: String(IAgentContextSizeService),
-    label: 'AgentContextSizeService',
+    id: String(IAgentTokenCountingService),
+    label: 'AgentTokenCountingService',
     scope: 'agent',
     fetch: (svc) => call(svc, 'get'),
   },

docs/en/configuration/config-files.md

@@ -252,6 +252,16 @@ When the experiment is enabled, the configuration is validated as the session st
 
 Retries only apply to transient failures — connection errors, timeouts, HTTP 429 rate limits, and 5xx server errors. A 429 caused by an exhausted quota or insufficient account balance is not retried and fails immediately, since it cannot succeed until the account is recharged.
 
+## `token_counting`
+
+`token_counting` selects which context token count is reported externally — the value behind the context-size display. Internal logic (automatic compaction triggers, budgets, and overflow backoff) always uses both provider-reported usage and estimates, regardless of this setting.
+
+| Field | Type | Default | Description |
+| --- | --- | --- | --- |
+| `strategy` | `"measured+estimated" \| "measured" \| "estimated"` | `"measured+estimated"` | `measured+estimated` reports the live size — the provider-reported usage of each exchange plus an estimate of the not-yet-measured tail — floored by the last measured total; `measured` reports provider usage alone, so the display only moves when an exchange completes; `estimated` reports a pure estimate with provider usage ignored — the fallback for providers that do not report usage or report it unreliably |
+
+`strategy` can be overridden by the `KIMI_TOKEN_COUNTING_STRATEGY` environment variable, which takes higher priority than `config.toml`.
+
 ## `background`
 
 `background` controls the concurrency behavior of background tasks (launched via the `Bash` tool or the `Agent` tool's `run_in_background=true` parameter).

docs/en/configuration/env-vars.md

@@ -135,6 +135,7 @@ Switches that control the behavior of subsystems such as telemetry, background t
 | `KIMI_MCP_TOOL_TIMEOUT_MS` | Global default single tool-call timeout (ms) for all MCP servers; takes higher priority than `[mcp] tool_timeout_ms` in `config.toml`, but a per-server `toolTimeoutMs` in `mcp.json` still wins (default `60000`) | Integer from `1` to `2147483647`; invalid values are ignored |
 | `KIMI_LOOP_MAX_STEPS_PER_TURN` | Maximum Agent steps per turn; takes higher priority than `[loop_control] max_steps_per_turn` in `config.toml` (unset or `0` means unlimited) | Non-negative integer; invalid values are ignored |
 | `KIMI_LOOP_MAX_ATTEMPTS_PER_STEP` | Maximum total attempts for a failing step (including the initial attempt); takes higher priority than `[loop_control] max_attempts_per_step` in `config.toml` (default `10`). The deprecated `KIMI_LOOP_MAX_RETRIES_PER_STEP` is still honored with a warning when this variable is unset | Non-negative integer; invalid values are ignored |
+| `KIMI_TOKEN_COUNTING_STRATEGY` | Which context token count is reported externally (the context-size display); takes higher priority than `[token_counting] strategy` in `config.toml` (default `measured+estimated`) | `measured+estimated`, `measured`, `estimated` (case-insensitive); invalid values are ignored |
 | `KIMI_WEB_SEARCH_BASE_URL` | API URL of the web search (`WebSearch`) service; takes higher priority than `[services.moonshot_search] base_url` in `config.toml`, and enables the service without that config section. Persisted credentials and custom headers are not forwarded to an env-selected endpoint | Non-blank string; blank values are ignored |
 | `KIMI_WEB_SEARCH_API_KEY` | API key of the web search (`WebSearch`) service; replaces both the configured API key and OAuth credential when set | Non-blank string; blank values are ignored |
 | `KIMI_WEB_FETCH_BASE_URL` | API URL of the web fetch (`FetchURL`) service; takes higher priority than `[services.moonshot_fetch] base_url`. Persisted credentials and custom headers are not forwarded to an env-selected endpoint. Without an env or config endpoint, signed-in users try the managed Kimi OAuth fetch service before direct local requests | Non-blank string; blank values are ignored |

docs/zh/configuration/config-files.md

@@ -252,6 +252,16 @@ max_output_size = 8192
 
 重试仅针对瞬时故障——连接错误、超时、HTTP 429 限流和 5xx 服务端错误。账户额度耗尽或余额不足导致的 429 不会重试，会立即失败：在充值之前重试不可能成功。
 
+## `token_counting`
+
+`token_counting` 决定对外上报的上下文 token 计数——即上下文大小显示所基于的值。内部逻辑（自动压缩触发、预算、超限退避）始终同时使用供应商实测与估算，不受本配置影响。
+
+| 字段 | 类型 | 默认值 | 说明 |
+| --- | --- | --- | --- |
+| `strategy` | `"measured+estimated" \| "measured" \| "estimated"` | `"measured+estimated"` | `measured+estimated` 上报实时大小——每次请求的供应商实测用量加上未实测尾部的估算——并以最近一次实测总量兜底；`measured` 只上报供应商实测，显示仅在每次请求完成后变化；`estimated` 忽略供应商实测、上报纯估算——适用于不上报用量或用量不可信的供应商 |
+
+`strategy` 可被环境变量 `KIMI_TOKEN_COUNTING_STRATEGY` 覆盖，优先级高于 `config.toml`。
+
 ## `background`
 
 `background` 控制后台任务（通过 `Bash` 工具或 `Agent` 工具的 `run_in_background=true` 参数启动）的并发数。

docs/zh/configuration/env-vars.md

@@ -135,6 +135,7 @@ kimi
 | `KIMI_MCP_TOOL_TIMEOUT_MS` | 所有 MCP server 的全局默认单次工具调用超时（毫秒）；优先级高于 `config.toml` 的 `[mcp] tool_timeout_ms`，但低于 `mcp.json` 中单个 server 的 `toolTimeoutMs`（默认 `60000`） | `1` 到 `2147483647` 的整数；非法值被忽略 |
 | `KIMI_LOOP_MAX_STEPS_PER_TURN` | Agent 单轮最大步数；优先级高于 `config.toml` 的 `[loop_control] max_steps_per_turn`（不设或 `0` 表示无上限） | 非负整数；非法值被忽略 |
 | `KIMI_LOOP_MAX_ATTEMPTS_PER_STEP` | 单步失败后的最大总尝试次数（含首次尝试）；优先级高于 `config.toml` 的 `[loop_control] max_attempts_per_step`（默认 `10`）。旧的 `KIMI_LOOP_MAX_RETRIES_PER_STEP` 已废弃，但在本变量未设置时仍生效并给出警告 | 非负整数；非法值被忽略 |
+| `KIMI_TOKEN_COUNTING_STRATEGY` | 对外上报的上下文 token 计数（上下文大小显示）；优先级高于 `config.toml` 的 `[token_counting] strategy`（默认 `measured+estimated`） | `measured+estimated`、`measured`、`estimated`（不区分大小写）；非法值被忽略 |
 | `KIMI_WEB_SEARCH_BASE_URL` | 网页搜索（`WebSearch`）服务的 API URL；优先级高于 `config.toml` 的 `[services.moonshot_search] base_url`，未写配置段时也可启用服务。文件中持久化的凭据和自定义 header 不会发送到环境变量指定的端点 | 非空字符串；空白值被忽略 |
 | `KIMI_WEB_SEARCH_API_KEY` | 网页搜索（`WebSearch`）服务的 API 密钥；设置后同时替换配置中的 API 密钥和 OAuth 凭据 | 非空字符串；空白值被忽略 |
 | `KIMI_WEB_FETCH_BASE_URL` | 网页抓取（`FetchURL`）服务的 API URL；优先级高于 `[services.moonshot_fetch] base_url`。文件中持久化的凭据和自定义 header 不会发送到环境变量指定的端点。环境变量和配置都没有指定端点时，已登录用户会先尝试 Kimi OAuth 托管抓取服务，再回退到本地直接请求 | 非空字符串；空白值被忽略 |

packages/agent-core-v2/docs/config-manifest.toml

@@ -8,7 +8,7 @@
 # commented "# field: type" lines describe the remaining schema fields.
 # Values resolve as: default -> config.toml -> env overlay -> memory.
 
-# Index (22 sections · 3 overlay(s))
+# Index (23 sections · 3 overlay(s))
 #   background               src/agent/task/configSection.ts
 #   cron                     src/app/cron/configSection.ts
 #   defaultPermissionMode    src/agent/permissionMode/configSection.ts
@@ -30,6 +30,7 @@
 #   subagent                 src/session/subagent/configSection.ts
 #   task                     src/agent/task/configSection.ts
 #   thinking                 src/app/kosongConfig/configSection.ts
+#   tokenCounting            src/agent/tokenCounting/configSection.ts
 #   tools                    src/agent/toolPolicy/configSection.ts
 #   (overlay) servicesCredentialEnvOverlay  src/app/auth/configSection.ts
 #   (overlay) kimiModelEnvOverlay  src/app/kosongConfig/envOverlay.ts
@@ -390,6 +391,18 @@ timeout_ms = 7200000
 # forced_effort: string
 # keep: string
 
+# ##########################################################################
+# tokenCounting (config.toml: token_counting)
+#   owner: src/agent/tokenCounting/configSection.ts
+#   scope: core
+#   hooks: stripEnv
+#   env:
+#     strategy <- KIMI_TOKEN_COUNTING_STRATEGY (custom parse)
+# ##########################################################################
+
+[token_counting]
+strategy = "measured+estimated"
+
 # ##########################################################################
 # tools
 #   owner: src/agent/toolPolicy/configSection.ts

packages/agent-core-v2/docs/state-manifest.d.ts

@@ -23,7 +23,7 @@
 // references become '(circular)', and class instances collapse to a '(ClassName)'
 // marker — the wire shape of an entry is the JSON projection of the type here.
 //
-// Index (App: 0 keys · Workspace: 6 keys · Session: 18 keys · Agent: 70 keys)
+// Index (App: 0 keys · Workspace: 6 keys · Session: 18 keys · Agent: 69 keys)
 //   App
 //   Workspace
 //     workspaceDirs.ephemeralDirs          src/workspace/workspaceDirs/workspaceDirsService.ts
@@ -62,7 +62,6 @@
 //     agentsMdReminder.seeded                         src/agent/agentsMdReminder/agentsMdReminderService.ts
 //     contextInjector.isNewTurn                       src/agent/contextInjector/contextInjectorService.ts
 //     contextProjector.lastRepairSignature            src/agent/contextProjector/contextProjectorService.ts
-//     contextSize.lastEmittedTokens                   src/agent/contextSize/contextSizeService.ts
 //     externalHooks.stopHookContinuationUsed          src/agent/externalHooks/externalHooksService.ts
 //     fullCompaction.activeTurnId                     src/agent/fullCompaction/fullCompactionService.ts
 //     fullCompaction.compactionCountInTurn            src/agent/fullCompaction/fullCompactionService.ts
@@ -981,8 +980,6 @@ export interface AgentStateSnapshot {
   'contextInjector.isNewTurn': boolean;
   // src/agent/contextProjector/contextProjectorService.ts
   'contextProjector.lastRepairSignature': string | null;
-  // src/agent/contextSize/contextSizeService.ts
-  'contextSize.lastEmittedTokens': number;
   // src/agent/externalHooks/externalHooksService.ts
   'externalHooks.stopHookContinuationUsed': boolean;
   // src/agent/fullCompaction/fullCompactionService.ts

packages/agent-core-v2/docs/wire-manifest.d.ts

@@ -21,9 +21,8 @@
 // owning model offloads inline media to blob storage), cross-reducers
 // (foreign models that also reduce this record on dispatch and replay).
 
-// Index (46 record types)
+// Index (48 record types)
 //   config.update                      profile               persisted  src/agent/profile/profileOps.ts
-//   context_size.measured              contextSize           transient  src/agent/contextSize/contextSizeOps.ts
 //   context.append_loop_event          contextMemory         persisted  src/agent/contextMemory/contextOps.ts
 //   context.append_message             contextMemory         persisted  src/agent/contextMemory/contextOps.ts
 //   context.apply_compaction           contextMemory         persisted  src/agent/contextMemory/contextOps.ts
@@ -58,6 +57,9 @@
 //   swarm_mode.exit                    swarm                 persisted  src/agent/swarm/swarmOps.ts
 //   task.started                       task                  persisted  src/agent/task/taskOps.ts
 //   task.terminated                    task                  persisted  src/agent/task/taskOps.ts
+//   token_counting.measured            tokenCounting         transient  src/agent/tokenCounting/tokenCountingOps.ts
+//   token_counting.rebased             tokenCounting         transient  src/agent/tokenCounting/tokenCountingOps.ts
+//   token_counting.truncated           tokenCounting         transient  src/agent/tokenCounting/tokenCountingOps.ts
 //   tools.register_user_tool           userTool              persisted  src/agent/userTool/userToolOps.ts
 //   tools.reset_active_tools           profile.activeTools   persisted  src/agent/profile/profileOps.ts
 //   tools.set_active_tools             profile.activeTools   persisted  src/agent/profile/profileOps.ts
@@ -86,16 +88,6 @@ interface ConfigUpdatePayload {
   disallowedTools?: string[];
 }
 
-/**
- * model: contextSize · toEvent
- * owner: src/agent/contextSize/contextSizeOps.ts
- */
-interface ContextSizeMeasuredPayload {
-  _name: 'context_size.measured';
-  length: number;
-  tokens: number;
-}
-
 /**
  * model: contextMemory · persisted · blobs · cross-reducers: turn
  * owner: src/agent/contextMemory/contextOps.ts
@@ -525,6 +517,37 @@ interface TaskTerminatedPayload {
   outputTail?: string;
 }
 
+/**
+ * model: tokenCounting · toEvent
+ * owner: src/agent/tokenCounting/tokenCountingOps.ts
+ */
+interface TokenCountingMeasuredPayload {
+  _name: 'token_counting.measured';
+  length: number;
+  tokens: number;
+}
+
+/**
+ * model: tokenCounting · toEvent
+ * owner: src/agent/tokenCounting/tokenCountingOps.ts
+ */
+interface TokenCountingRebasedPayload {
+  _name: 'token_counting.rebased';
+  length: number;
+  tokens: number;
+  measured: boolean;
+}
+
+/**
+ * model: tokenCounting · toEvent
+ * owner: src/agent/tokenCounting/tokenCountingOps.ts
+ */
+interface TokenCountingTruncatedPayload {
+  _name: 'token_counting.truncated';
+  length: number;
+  tokens: number;
+}
+
 /**
  * model: userTool · persisted
  * owner: src/agent/userTool/userToolOps.ts
@@ -683,7 +706,6 @@ interface UsageRecordPayload {
 /** Record type → payload sketch. */
 interface WirePayloadMap {
   "config.update": ConfigUpdatePayload;
-  "context_size.measured": ContextSizeMeasuredPayload;
   "context.append_loop_event": ContextAppendLoopEventPayload;
   "context.append_message": ContextAppendMessagePayload;
   "context.apply_compaction": ContextApplyCompactionPayload;
@@ -718,6 +740,9 @@ interface WirePayloadMap {
   "swarm_mode.exit": SwarmModeExitPayload;
   "task.started": TaskStartedPayload;
   "task.terminated": TaskTerminatedPayload;
+  "token_counting.measured": TokenCountingMeasuredPayload;
+  "token_counting.rebased": TokenCountingRebasedPayload;
+  "token_counting.truncated": TokenCountingTruncatedPayload;
   "tools.register_user_tool": ToolsRegisterUserToolPayload;
   "tools.reset_active_tools": ToolsResetActiveToolsPayload;
   "tools.set_active_tools": ToolsSetActiveToolsPayload;