openclaw/docs/zh-CN/pi.md

title, x-i18n

title	x-i18n
Pi 集成架构	generated_at model provider source_hash source_path workflow 2026-02-03T07:53:24Z claude-opus-4-5 pi 98b12f1211f70b1a25f58e68c7a4d0fe3827412ca53ba0ea2cd41ac9c0448458 pi.md 15

Pi 集成架构

本文档描述了 OpenClaw 如何与 pi-coding-agent 及其相关包（pi-ai、pi-agent-core、pi-tui）集成以实现其 AI 智能体能力。

概述

OpenClaw 使用 pi SDK 将 AI 编码智能体嵌入到其消息 Gateway 网关架构中。OpenClaw 不是将 pi 作为子进程生成或使用 RPC 模式，而是通过 createAgentSession() 直接导入并实例化 pi 的 AgentSession。这种嵌入式方法提供了：

• 对会话生命周期和事件处理的完全控制
• 自定义工具注入（消息、沙箱、渠道特定操作）
• 每个渠道/上下文的系统提示自定义
• 支持分支/压缩的会话持久化
• 带故障转移的多账户认证配置文件轮换
• 与提供商无关的模型切换

包依赖

{
  "@mariozechner/pi-agent-core": "0.49.3",
  "@mariozechner/pi-ai": "0.49.3",
  "@mariozechner/pi-coding-agent": "0.49.3",
  "@mariozechner/pi-tui": "0.49.3"
}

包	用途
pi-ai	核心 LLM 抽象：Model、streamSimple、消息类型、提供商 API
pi-agent-core	智能体循环、工具执行、AgentMessage 类型
pi-coding-agent	高级 SDK：createAgentSession、SessionManager、AuthStorage、ModelRegistry、内置工具
pi-tui	终端 UI 组件（用于 OpenClaw 的本地 TUI 模式）

文件结构

src/agents/
├── pi-embedded-runner.ts          # Re-exports from pi-embedded-runner/
├── pi-embedded-runner/
│   ├── run.ts                     # Main entry: runEmbeddedPiAgent()
│   ├── run/
│   │   ├── attempt.ts             # Single attempt logic with session setup
│   │   ├── params.ts              # RunEmbeddedPiAgentParams type
│   │   ├── payloads.ts            # Build response payloads from run results
│   │   ├── images.ts              # Vision model image injection
│   │   └── types.ts               # EmbeddedRunAttemptResult
│   ├── abort.ts                   # Abort error detection
│   ├── cache-ttl.ts               # Cache TTL tracking for context pruning
│   ├── compact.ts                 # Manual/auto compaction logic
│   ├── extensions.ts              # Load pi extensions for embedded runs
│   ├── extra-params.ts            # Provider-specific stream params
│   ├── google.ts                  # Google/Gemini turn ordering fixes
│   ├── history.ts                 # History limiting (DM vs group)
│   ├── lanes.ts                   # Session/global command lanes
│   ├── logger.ts                  # Subsystem logger
│   ├── model.ts                   # Model resolution via ModelRegistry
│   ├── runs.ts                    # Active run tracking, abort, queue
│   ├── sandbox-info.ts            # Sandbox info for system prompt
│   ├── session-manager-cache.ts   # SessionManager instance caching
│   ├── session-manager-init.ts    # Session file initialization
│   ├── system-prompt.ts           # System prompt builder
│   ├── tool-split.ts              # Split tools into builtIn vs custom
│   ├── types.ts                   # EmbeddedPiAgentMeta, EmbeddedPiRunResult
│   └── utils.ts                   # ThinkLevel mapping, error description
├── pi-embedded-subscribe.ts       # Session event subscription/dispatch
├── pi-embedded-subscribe.types.ts # SubscribeEmbeddedPiSessionParams
├── pi-embedded-subscribe.handlers.ts # Event handler factory
├── pi-embedded-subscribe.handlers.lifecycle.ts
├── pi-embedded-subscribe.handlers.types.ts
├── pi-embedded-block-chunker.ts   # Streaming block reply chunking
├── pi-embedded-messaging.ts       # Messaging tool sent tracking
├── pi-embedded-helpers.ts         # Error classification, turn validation
├── pi-embedded-helpers/           # Helper modules
├── pi-embedded-utils.ts           # Formatting utilities
├── pi-tools.ts                    # createOpenClawCodingTools()
├── pi-tools.abort.ts              # AbortSignal wrapping for tools
├── pi-tools.policy.ts             # Tool allowlist/denylist policy
├── pi-tools.read.ts               # Read tool customizations
├── pi-tools.schema.ts             # Tool schema normalization
├── pi-tools.types.ts              # AnyAgentTool type alias
├── pi-tool-definition-adapter.ts  # AgentTool -> ToolDefinition adapter
├── pi-settings.ts                 # Settings overrides
├── pi-extensions/                 # Custom pi extensions
│   ├── compaction-safeguard.ts    # Safeguard extension
│   ├── compaction-safeguard-runtime.ts
│   ├── context-pruning.ts         # Cache-TTL context pruning extension
│   └── context-pruning/
├── model-auth.ts                  # Auth profile resolution
├── auth-profiles.ts               # Profile store, cooldown, failover
├── model-selection.ts             # Default model resolution
├── models-config.ts               # models.json generation
├── model-catalog.ts               # Model catalog cache
├── context-window-guard.ts        # Context window validation
├── failover-error.ts              # FailoverError class
├── defaults.ts                    # DEFAULT_PROVIDER, DEFAULT_MODEL
├── system-prompt.ts               # buildAgentSystemPrompt()
├── system-prompt-params.ts        # System prompt parameter resolution
├── system-prompt-report.ts        # Debug report generation
├── tool-summaries.ts              # Tool description summaries
├── tool-policy.ts                 # Tool policy resolution
├── transcript-policy.ts           # Transcript validation policy
├── skills.ts                      # Skill snapshot/prompt building
├── skills/                        # Skill subsystem
├── sandbox.ts                     # Sandbox context resolution
├── sandbox/                       # Sandbox subsystem
├── channel-tools.ts               # Channel-specific tool injection
├── openclaw-tools.ts              # OpenClaw-specific tools
├── bash-tools.ts                  # exec/process tools
├── apply-patch.ts                 # apply_patch tool (OpenAI)
├── tools/                         # Individual tool implementations
│   ├── browser-tool.ts
│   ├── canvas-tool.ts
│   ├── cron-tool.ts
│   ├── discord-actions*.ts
│   ├── gateway-tool.ts
│   ├── image-tool.ts
│   ├── message-tool.ts
│   ├── nodes-tool.ts
│   ├── session*.ts
│   ├── slack-actions.ts
│   ├── telegram-actions.ts
│   ├── web-*.ts
│   └── whatsapp-actions.ts
└── ...

核心集成流程

1. 运行嵌入式智能体

主入口点是 pi-embedded-runner/run.ts 中的 runEmbeddedPiAgent()：

import { runEmbeddedPiAgent } from "./agents/pi-embedded-runner.js";

const result = await runEmbeddedPiAgent({
  sessionId: "user-123",
  sessionKey: "main:whatsapp:+1234567890",
  sessionFile: "/path/to/session.jsonl",
  workspaceDir: "/path/to/workspace",
  config: openclawConfig,
  prompt: "Hello, how are you?",
  provider: "anthropic",
  model: "claude-sonnet-4-20250514",
  timeoutMs: 120_000,
  runId: "run-abc",
  onBlockReply: async (payload) => {
    await sendToChannel(payload.text, payload.mediaUrls);
  },
});

2. 会话创建

在 runEmbeddedAttempt()（由 runEmbeddedPiAgent() 调用）内部，使用 pi SDK：

import {
  createAgentSession,
  DefaultResourceLoader,
  SessionManager,
  SettingsManager,
} from "@mariozechner/pi-coding-agent";

const resourceLoader = new DefaultResourceLoader({
  cwd: resolvedWorkspace,
  agentDir,
  settingsManager,
  additionalExtensionPaths,
});
await resourceLoader.reload();

const { session } = await createAgentSession({
  cwd: resolvedWorkspace,
  agentDir,
  authStorage: params.authStorage,
  modelRegistry: params.modelRegistry,
  model: params.model,
  thinkingLevel: mapThinkingLevel(params.thinkLevel),
  tools: builtInTools,
  customTools: allCustomTools,
  sessionManager,
  settingsManager,
  resourceLoader,
});

applySystemPromptOverrideToSession(session, systemPromptOverride);

3. 事件订阅

subscribeEmbeddedPiSession() 订阅 pi 的 AgentSession 事件：

const subscription = subscribeEmbeddedPiSession({
  session: activeSession,
  runId: params.runId,
  verboseLevel: params.verboseLevel,
  reasoningMode: params.reasoningLevel,
  toolResultFormat: params.toolResultFormat,
  onToolResult: params.onToolResult,
  onReasoningStream: params.onReasoningStream,
  onBlockReply: params.onBlockReply,
  onPartialReply: params.onPartialReply,
  onAgentEvent: params.onAgentEvent,
});

处理的事件包括：

• message_start / message_end / message_update（流式文本/思考）
• tool_execution_start / tool_execution_update / tool_execution_end
• turn_start / turn_end
• agent_start / agent_end
• auto_compaction_start / auto_compaction_end

4. 提示

设置完成后，会话被提示：

await session.prompt(effectivePrompt, { images: imageResult.images });

SDK 处理完整的智能体循环：发送到 LLM、执行工具调用、流式响应。

工具架构

工具管道

1. 基础工具：pi 的 codingTools（read、bash、edit、write）
2. 自定义替换：OpenClaw 将 bash 替换为 exec/process，为沙箱自定义 read/edit/write
3. OpenClaw 工具：消息、浏览器、画布、会话、定时任务、Gateway 网关等
4. 渠道工具：Discord/Telegram/Slack/WhatsApp 特定的操作工具
5. 策略过滤：工具按配置文件、提供商、智能体、群组、沙箱策略过滤
6. Schema 规范化：为 Gemini/OpenAI 的特殊情况清理 Schema
7. AbortSignal 包装：工具被包装以尊重中止信号

工具定义适配器

pi-agent-core 的 AgentTool 与 pi-coding-agent 的 ToolDefinition 有不同的 execute 签名。pi-tool-definition-adapter.ts 中的适配器桥接了这一点：

export function toToolDefinitions(tools: AnyAgentTool[]): ToolDefinition[] {
  return tools.map((tool) => ({
    name: tool.name,
    label: tool.label ?? name,
    description: tool.description ?? "",
    parameters: tool.parameters,
    execute: async (toolCallId, params, onUpdate, _ctx, signal) => {
      // pi-coding-agent signature differs from pi-agent-core
      return await tool.execute(toolCallId, params, signal, onUpdate);
    },
  }));
}

工具拆分策略

splitSdkTools() 通过 customTools 传递所有工具：

export function splitSdkTools(options: { tools: AnyAgentTool[]; sandboxEnabled: boolean }) {
  return {
    builtInTools: [], // Empty. We override everything
    customTools: toToolDefinitions(options.tools),
  };
}

这确保 OpenClaw 的策略过滤、沙箱集成和扩展工具集在各提供商之间保持一致。

系统提示构建

系统提示在 buildAgentSystemPrompt()（system-prompt.ts）中构建。它组装一个完整的提示，包含工具、工具调用风格、安全护栏、OpenClaw CLI 参考、Skills、文档、工作区、沙箱、消息、回复标签、语音、静默回复、心跳、运行时元数据等部分，以及启用时的记忆和反应，还有可选的上下文文件和额外系统提示内容。部分内容在子智能体使用的最小提示模式下会被裁剪。

提示在会话创建后通过 applySystemPromptOverrideToSession() 应用：

const systemPromptOverride = createSystemPromptOverride(appendPrompt);
applySystemPromptOverrideToSession(session, systemPromptOverride);

会话管理

会话文件

会话是具有树结构（id/parentId 链接）的 JSONL 文件。Pi 的 SessionManager 处理持久化：

const sessionManager = SessionManager.open(params.sessionFile);

OpenClaw 用 guardSessionManager() 包装它以确保工具结果安全。

会话缓存

session-manager-cache.ts 缓存 SessionManager 实例以避免重复的文件解析：

await prewarmSessionFile(params.sessionFile);
sessionManager = SessionManager.open(params.sessionFile);
trackSessionManagerAccess(params.sessionFile);

历史限制

limitHistoryTurns() 根据渠道类型（私信 vs 群组）裁剪对话历史。

压缩

自动压缩在上下文溢出时触发。compactEmbeddedPiSessionDirect() 处理手动压缩：

const compactResult = await compactEmbeddedPiSessionDirect({
  sessionId, sessionFile, provider, model, ...
});

认证与模型解析

认证配置文件

OpenClaw 维护一个认证配置文件存储，每个提供商有多个 API 密钥：

const authStore = ensureAuthProfileStore(agentDir, { allowKeychainPrompt: false });
const profileOrder = resolveAuthProfileOrder({ cfg, store: authStore, provider, preferredProfile });

配置文件在失败时轮换，并带有冷却跟踪：

await markAuthProfileFailure({ store, profileId, reason, cfg, agentDir });
const rotated = await advanceAuthProfile();

模型解析

import { resolveModel } from "./pi-embedded-runner/model.js";

const { model, error, authStorage, modelRegistry } = resolveModel(
  provider,
  modelId,
  agentDir,
  config,
);

// Uses pi's ModelRegistry and AuthStorage
authStorage.setRuntimeApiKey(model.provider, apiKeyInfo.apiKey);

故障转移

FailoverError 在配置了回退时触发模型回退：

if (fallbackConfigured && isFailoverErrorMessage(errorText)) {
  throw new FailoverError(errorText, {
    reason: promptFailoverReason ?? "unknown",
    provider,
    model: modelId,
    profileId,
    status: resolveFailoverStatus(promptFailoverReason),
  });
}

Pi 扩展

OpenClaw 加载自定义 pi 扩展以实现特殊行为：

压缩安全护栏

pi-extensions/compaction-safeguard.ts 为压缩添加护栏，包括自适应令牌预算以及工具失败和文件操作摘要：

if (resolveCompactionMode(params.cfg) === "safeguard") {
  setCompactionSafeguardRuntime(params.sessionManager, { maxHistoryShare });
  paths.push(resolvePiExtensionPath("compaction-safeguard"));
}

上下文裁剪

pi-extensions/context-pruning.ts 实现基于缓存 TTL 的上下文裁剪：

if (cfg?.agents?.defaults?.contextPruning?.mode === "cache-ttl") {
  setContextPruningRuntime(params.sessionManager, {
    settings,
    contextWindowTokens,
    isToolPrunable,
    lastCacheTouchAt,
  });
  paths.push(resolvePiExtensionPath("context-pruning"));
}

流式传输与块回复

块分块

EmbeddedBlockChunker 管理将流式文本分成离散的回复块：

const blockChunker = blockChunking ? new EmbeddedBlockChunker(blockChunking) : null;

思考/最终标签剥离

流式输出被处理以剥离 <think>/<thinking> 块并提取 <final> 内容：

const stripBlockTags = (text: string, state: { thinking: boolean; final: boolean }) => {
  // Strip <think>...</think> content
  // If enforceFinalTag, only return <final>...</final> content
};

回复指令

回复指令如 [[media:url]]、[[voice]]、[[reply:id]] 被解析和提取：

const { text: cleanedText, mediaUrls, audioAsVoice, replyToId } = consumeReplyDirectives(chunk);

错误处理

错误分类

pi-embedded-helpers.ts 对错误进行分类以进行适当处理：

isContextOverflowError(errorText)     // Context too large
isCompactionFailureError(errorText)   // Compaction failed
isAuthAssistantError(lastAssistant)   // Auth failure
isRateLimitAssistantError(...)        // Rate limited
isFailoverAssistantError(...)         // Should failover
classifyFailoverReason(errorText)     // "auth" | "rate_limit" | "quota" | "timeout" | ...

思考级别回退

如果思考级别不受支持，它会回退：

const fallbackThinking = pickFallbackThinkingLevel({
  message: errorText,
  attempted: attemptedThinking,
});
if (fallbackThinking) {
  thinkLevel = fallbackThinking;
  continue;
}

沙箱集成

当启用沙箱模式时，工具和路径受到约束：

const sandbox = await resolveSandboxContext({
  config: params.config,
  sessionKey: sandboxSessionKey,
  workspaceDir: resolvedWorkspace,
});

if (sandboxRoot) {
  // Use sandboxed read/edit/write tools
  // Exec runs in container
  // Browser uses bridge URL
}

提供商特定处理

Anthropic

• 拒绝魔术字符串清除
• 连续角色的回合验证
• Claude Code 参数兼容性

Google/Gemini

• 回合排序修复（applyGoogleTurnOrderingFix）
• 工具 schema 清理（sanitizeToolsForGoogle）
• 会话历史清理（sanitizeSessionHistory）

OpenAI

• Codex 模型的 apply_patch 工具
• 思考级别降级处理

TUI 集成

OpenClaw 还有一个本地 TUI 模式，直接使用 pi-tui 组件：

// src/tui/tui.ts
import { ... } from "@mariozechner/pi-tui";

这提供了与 pi 原生模式类似的交互式终端体验。

与 Pi CLI 的主要区别

方面	Pi CLI	OpenClaw 嵌入式
调用方式	pi 命令 / RPC	通过 createAgentSession() 的 SDK
工具	默认编码工具	自定义 OpenClaw 工具套件
系统提示	AGENTS.md + prompts	按渠道/上下文动态生成
会话存储	~/.pi/agent/sessions/	~/.openclaw/agents/<agentId>/sessions/（或 $OPENCLAW_STATE_DIR/agents/<agentId>/sessions/）
认证	单一凭证	带轮换的多配置文件
扩展	从磁盘加载	编程方式 + 磁盘路径
事件处理	TUI 渲染	基于回调（onBlockReply 等）

未来考虑

可能需要重构的领域：

1. 工具签名对齐：目前在 pi-agent-core 和 pi-coding-agent 签名之间适配
2. 会话管理器包装：guardSessionManager 增加了安全性但增加了复杂性
3. 扩展加载：可以更直接地使用 pi 的 ResourceLoader
4. 流式处理器复杂性：subscribeEmbeddedPiSession 已经变得很大
5. 提供商特殊情况：许多提供商特定的代码路径，pi 可能可以处理

测试

所有涵盖 pi 集成及其扩展的现有测试：

• src/agents/pi-embedded-block-chunker.test.ts
• src/agents/pi-embedded-helpers.buildbootstrapcontextfiles.test.ts
• src/agents/pi-embedded-helpers.classifyfailoverreason.test.ts
• src/agents/pi-embedded-helpers.downgradeopenai-reasoning.test.ts
• src/agents/pi-embedded-helpers.formatassistanterrortext.test.ts
• src/agents/pi-embedded-helpers.formatrawassistanterrorforui.test.ts
• src/agents/pi-embedded-helpers.image-dimension-error.test.ts
• src/agents/pi-embedded-helpers.image-size-error.test.ts
• src/agents/pi-embedded-helpers.isautherrormessage.test.ts
• src/agents/pi-embedded-helpers.isbillingerrormessage.test.ts
• src/agents/pi-embedded-helpers.iscloudcodeassistformaterror.test.ts
• src/agents/pi-embedded-helpers.iscompactionfailureerror.test.ts
• src/agents/pi-embedded-helpers.iscontextoverflowerror.test.ts
• src/agents/pi-embedded-helpers.isfailovererrormessage.test.ts
• src/agents/pi-embedded-helpers.islikelycontextoverflowerror.test.ts
• src/agents/pi-embedded-helpers.ismessagingtoolduplicate.test.ts
• src/agents/pi-embedded-helpers.messaging-duplicate.test.ts
• src/agents/pi-embedded-helpers.normalizetextforcomparison.test.ts
• src/agents/pi-embedded-helpers.resolvebootstrapmaxchars.test.ts
• src/agents/pi-embedded-helpers.sanitize-session-messages-images.keeps-tool-call-tool-result-ids-unchanged.test.ts
• src/agents/pi-embedded-helpers.sanitize-session-messages-images.removes-empty-assistant-text-blocks-but-preserves.test.ts
• src/agents/pi-embedded-helpers.sanitizegoogleturnordering.test.ts
• src/agents/pi-embedded-helpers.sanitizesessionmessagesimages-thought-signature-stripping.test.ts
• src/agents/pi-embedded-helpers.sanitizetoolcallid.test.ts
• src/agents/pi-embedded-helpers.sanitizeuserfacingtext.test.ts
• src/agents/pi-embedded-helpers.stripthoughtsignatures.test.ts
• src/agents/pi-embedded-helpers.validate-turns.test.ts
• src/agents/pi-embedded-runner-extraparams.live.test.ts（实时）
• src/agents/pi-embedded-runner-extraparams.test.ts
• src/agents/pi-embedded-runner.applygoogleturnorderingfix.test.ts
• src/agents/pi-embedded-runner.buildembeddedsandboxinfo.test.ts
• src/agents/pi-embedded-runner.createsystempromptoverride.test.ts
• src/agents/pi-embedded-runner.get-dm-history-limit-from-session-key.falls-back-provider-default-per-dm-not.test.ts
• src/agents/pi-embedded-runner.get-dm-history-limit-from-session-key.returns-undefined-sessionkey-is-undefined.test.ts
• src/agents/pi-embedded-runner.google-sanitize-thinking.test.ts
• src/agents/pi-embedded-runner.guard.test.ts
• src/agents/pi-embedded-runner.limithistoryturns.test.ts
• src/agents/pi-embedded-runner.resolvesessionagentids.test.ts
• src/agents/pi-embedded-runner.run-embedded-pi-agent.auth-profile-rotation.test.ts
• src/agents/pi-embedded-runner.sanitize-session-history.test.ts
• src/agents/pi-embedded-runner.splitsdktools.test.ts
• src/agents/pi-embedded-runner.test.ts
• src/agents/pi-embedded-subscribe.code-span-awareness.test.ts
• src/agents/pi-embedded-subscribe.reply-tags.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.calls-onblockreplyflush-before-tool-execution-start-preserve.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.does-not-append-text-end-content-is.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.does-not-call-onblockreplyflush-callback-is-not.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.does-not-duplicate-text-end-repeats-full.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.does-not-emit-duplicate-block-replies-text.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.emits-block-replies-text-end-does-not.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.emits-reasoning-as-separate-message-enabled.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.filters-final-suppresses-output-without-start-tag.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.includes-canvas-action-metadata-tool-summaries.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.keeps-assistanttexts-final-answer-block-replies-are.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.keeps-indented-fenced-blocks-intact.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.reopens-fenced-blocks-splitting-inside-them.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.splits-long-single-line-fenced-blocks-reopen.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.streams-soft-chunks-paragraph-preference.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.subscribeembeddedpisession.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.suppresses-message-end-block-replies-message-tool.test.ts
• src/agents/pi-embedded-subscribe.subscribe-embedded-pi-session.waits-multiple-compaction-retries-before-resolving.test.ts
• src/agents/pi-embedded-subscribe.tools.test.ts
• src/agents/pi-embedded-utils.test.ts
• src/agents/pi-extensions/compaction-safeguard.test.ts
• src/agents/pi-extensions/context-pruning.test.ts
• src/agents/pi-settings.test.ts
• src/agents/pi-tool-definition-adapter.test.ts
• src/agents/pi-tools-agent-config.test.ts
• src/agents/pi-tools.create-openclaw-coding-tools.adds-claude-style-aliases-schemas-without-dropping-b.test.ts
• src/agents/pi-tools.create-openclaw-coding-tools.adds-claude-style-aliases-schemas-without-dropping-d.test.ts
• src/agents/pi-tools.create-openclaw-coding-tools.adds-claude-style-aliases-schemas-without-dropping-f.test.ts
• src/agents/pi-tools.create-openclaw-coding-tools.adds-claude-style-aliases-schemas-without-dropping.test.ts
• src/agents/pi-tools.policy.test.ts
• src/agents/pi-tools.safe-bins.test.ts
• src/agents/pi-tools.workspace-paths.test.ts