clawd-code - This repository hosts a static mirror of a TypeScr

项目说明

Clawd Code — security research mirror

This repository hosts a static mirror of a TypeScript src/ tree for educational use, defensi research, and software supply-chain analysis. It is maintained by an independent uni student as a personal research project.

Naming. Splinterworks, Clawd, and Clawd Code are fictional stand-ins for a , assistant, and terminal coding-agent CLI—same spirit as Macrohard for a giant software company. They name a _style_ of product for this mirror, not a real shipping brand.

---

Important notice

Please read this section before using or redistributing anything from this repository.

---

Purpose and acceptable use

This archi to support:

Use it only in ways that comply with applicable law and third-party terms. Do not use this repository to harm others, bypass security, or misuse upstream services.

---

Background (high level)

Public discussion in March 2026 suggested that source map artifacts associated with an npm distribution may have made TypeScript source material easier to obtain than intended.

What is in this repository

The tree contains a mirrored src/ snapshot preserved for research and analysis.

Clawd Code (upstream) is a terminal-oriented CLI used with Splinterworks’s coding-assistant stack for tasks such as editing files, running commands, searching codebases, and coordinating workflows. This mirror is not a supported build, fork, or replacement for the official product.

| Attribute | Detail |
| ---------------------- | -------------------------------------------------- |
| Snapshot context | Public discussion, 2026-03-31 |
| Language | TypeScript |
| Runtime (upstream) | Bun |
| Terminal UI (upstream) | React + Ink |
| Scale (approximate) | ~1,900 files, 512,000+ lines of code |

---

Documentation ()

Na documentation for the src/ tree li docs/. It mirrors and expands this README into a sidebar (guide, architecture, reference) and includes a generated index of top-level src/ entries.

From the repository root:

---

Takedown and contact

If you are a rights holder or authorized representative and belie in this repository infringes your rights or applicable law, contact the maintainer by opening a GitHub Issue on this repository (or use any contact method the maintainer publishes on their GitHub profile).

Upon receipt of a valid legal notice (for example, a DMCA notice that meets statutory requirements where applicable), the maintainer will review it promptly and, where appropriate, disable or remove the identified content or the repository. Good-faith cooperation does not waive any party’s legal rights or remedies.

---

Directory structure

---

Architecture summary

The following describes obser in the mirrored tree for research purposes. It is descripti, not official product documentation.

1. Tool system (src/tools/)

Each in tool is implemented as a self-contained module defining input schema, permission model, and execution logic.

| Tool | Description |
| ---------------------------------------- | ---------------------------------------------- |
| BashTool | Shell command execution |
| FileReadTool | File reading (images, PDFs, notebooks) |
| FileWriteTool | File creation / o |
| FileEditTool | Partial file modification (string replacement) |
| GlobTool | File pattern matching search |
| GrepTool | ripgrep-based content search |
| WebFetchTool | Fetch URL content |
| WebSearchTool | Web search |
| AgentTool | Sub-agent spawning |
| SkillTool | Skill execution |
| MCPTool | MCP server tool in |
| LSPTool | Language Ser integration |
| NotebookEditTool | Jupyter notebook editing |
| TaskCreateTool / TaskUpdateTool | Task creation and management |
| SendMessageTool | Inter-agent messaging |
| TeamCreateTool / TeamDeleteTool | Team agent management |
| EnterPlanModeTool / ExitPlanModeTool | Plan mode toggle |
| EnterWorktreeTool / ExitWorktreeTool | Git worktree isolation |
| ToolSearchTool | Deferred tool discovery |
| CronCreateTool | Scheduled trigger creation |
| RemoteTriggerTool | Remote trigger |
| SleepTool | Proactive mode wait |
| SyntheticOutputTool | Structured output generation |

2. Command system (src/commands/)

User-facing slash commands invoked with a / prefix.

| Command | Description |
| -------------------- | ---------------------------- |
| /commit | Create a git commit |
| /review | Code review |
| /compact | Context compression |
| /mcp | MCP ser |
| /config | Settings management |
| /doctor | En diagnostics |
| /login / /logout | Authentication |
| /memory | Persistent memory management |
| /skills | Skill management |
| /tasks | Task management |
| /vim | Vim mode toggle |
| /diff | View changes |
| /cost | Check usage cost |
| /theme | Change theme |
| /context | Context |
| /pr_comments | View PR comments |
| /resume | Restore previous session |
| /share | Share session |
| /desktop | Desktop app handoff |
| /mobile | Mobile app handoff |

3. Service layer (src/services/)

| Service | Description |
| ------------------------ | ------------------------------------------------------- |
| api/ | Splinterworks API client, file API, bootstrap |
| mcp/ | Model Context Protocol ser and management |
| oauth/ | OAuth 2.0 authentication flow |
| lsp/ | Language Ser manager |
| analytics/ | GrowthBook-based feature flags and analytics |
| plugins/ | Plugin loader |
| compact/ | Con context compression |
| policyLimits/ | Organization policy limits |
| remoteManagedSettings/ | Remote managed settings |
| extractMemories/ | Automatic memory extraction |
| tokenEstimation.ts | Token count estimation |
| teamMemorySync/ | Team memory synchronization |

4. Bridge system (src/bridge/)

Bidirectional communication between IDE extensions (VS Code, JetBrains) and the CLI.

5. Permission system (src/hooks/toolPermission/)

Permission checks on tool in: user prompt or automatic resolution depending on configured permission mode (default, plan, bypassPermissions, auto, etc.).

6. Feature flags

Dead code elimination via Bun’s bun:bundle feature flags:

Notable flags: PROACTIVE, KAIROS, BRIDGE_MODE, DAEMON, , AGENT_TRIGGERS, MONITOR_TOOL

---

Key files in detail

QueryEngine.ts (~46K lines)

Core engine for LLM API calls: streaming responses, tool-call loops, thinking mode, retry logic, token counting.

Tool.ts (~29K lines)

Base types and interfaces for tools — input schemas, permission models, progress state types.

commands.ts (~25K lines)

Registration and execution of slash commands; conditional imports for en command sets.

main.tsx

Commander.js-based CLI parsing and React/Ink renderer setup; startup paths include MDM settings, keychain prefetch, and GrowthBook initialization for boot performance.

---

Tech stack

| Category | Technology |
| ----------------- | ----------------------------------------------------------------------- |
| Runtime | Bun |
| Language | TypeScript (strict) |
| Terminal UI | React + Ink |
| CLI Parsing | Commander.js (extra-typings) |
| Schema | Zod v4 |
| Code Search | ripgrep |
| Protocols | MCP SDK, LSP |
| API | Splinterworks SDK |
| Telemetry | OpenTelemetry + gRPC |
| Feature Flags | GrowthBook |
| Auth | OAuth 2.0, JWT, macOS Keychain |

---

Notable design patterns

Parallel prefetch

Startup prefetch of MDM settings, keychain reads, and API preconnect in parallel before heavier module e.

Lazy loading

Hea (OpenTelemetry, gRPC, analytics, and some feature-gated subsystems) deferred import() until needed.

Agent swarms

Sub-agents via AgentTool; coordinator/ for multi-agent orchestration; TeamCreateTool for team-le work.

Skill system

Workflows under skills/ executed through SkillTool; upstream product supports user-defined skills.

Plugin architecture

Plugins loaded through the plugins/ subsystem.

---

Footer

Non-affiliation. Not affiliated with, endorsed by, or maintained by Splinterworks or any named rights holder.

Intellectual property. Original works remain the property of their respecti; this mirror does not transfer any rights.

Disclaimer. Provided “AS IS” without warranty; not legal advice. Use at your own risk.

项目信息

开发语言：TypeScript

Star 数：35

技术要点

该项目属于防护思路方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕This repository hosts a static mirror of a TypeScript src/ tree for educational use, defensive security research, and software supply-chain analysis. It is maintained by an independent university student as a personal research project.展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。