← Back to Skills Marketplace
96
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install code-explore
Description
Analyze local codebase to explain directory structure, entrypoints, dependencies, call chains, and data flow without modifying code.
README (SKILL.md)
项目探索
先把代码库看清楚，再谈修改。
工作流
- 先看与需求相关的目录结构、依赖文件、配置文件和构建入口。
- 确认运行时、框架、主入口和关键启动路径。
- 只基于已经读过的文件梳理 import、export、调用链或事件流。
- 找出状态、模型、存储、外部服务和配置边界。
- 结论必须有代码依据；没读到的部分就明确写“不确定”。
- 使用本技能时不改代码，只做分析。
关注点
- 与当前问题直接相关的目录结构
- 入口文件和关键配置
- 核心模块及其职责
- 关键函数、处理器与调用关系
- 从输入到副作用或输出的数据流
输出
- 相关目录结构
- 入口与配置摘要
- 核心模块说明
- 关键函数和调用关系
- 简短的数据流说明
- 未确认项与假设
Usage Guidance
This skill appears coherent and only asks the agent to read and analyze code; it does not request credentials or install software. Before enabling: (1) confirm the agent will only have read access to the target repository and not broader host files, (2) avoid running it on repos that contain secrets or credentials (remove or mask sensitive files first), and (3) review any analysis outputs before sharing them, since they could include snippets from the codebase. If you need stricter controls, run the skill against a sanitized copy of the project or restrict the agent's file access scope.
Capability Analysis
Type: OpenClaw Skill
Name: code-explore
Version: 0.1.0
The 'code-explore' skill bundle is designed for static codebase analysis, focusing on directory structures, dependencies, and data flow. The instructions in SKILL.md explicitly restrict the agent to analysis only, forbidding code modifications or PR operations, and there are no indicators of malicious execution, data exfiltration, or prompt injection.
Capability Assessment
Purpose & Capability
Name and description match the SKILL.md: the skill's goal is to inspect a local codebase and explain structure, entry points, dependencies, call paths, and data flow. It does not declare unrelated binaries, env vars, or install steps.
Instruction Scope
SKILL.md explicitly limits actions to reading/analyzing code and states 'do not change code.' It focuses on reading repo files (directory structure, imports/exports, call chains) and marking unknowns. It does not instruct the agent to read unrelated system files, credentials, or to transmit data externally.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to execute. Nothing is written to disk or downloaded by the skill itself.
Credentials
The skill requests no environment variables, credentials, or config paths. Its requested access (read-only access to the repository files) is proportionate to its stated purpose. Note: while the skill itself doesn't request secrets, analysis of a repo that contains secrets could reveal them in outputs.
Persistence & Privilege
always is false and there are no install hooks or behavior that would modify agent/system configuration. The skill does not request permanent presence or permissions beyond runtime read access to the codebase.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install code-explore - After installation, invoke the skill by name or use
/code-explore - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
initial release
Metadata
Frequently Asked Questions
What is Code Explore?
Analyze local codebase to explain directory structure, entrypoints, dependencies, call chains, and data flow without modifying code. It is an AI Agent Skill for Claude Code / OpenClaw, with 96 downloads so far.
How do I install Code Explore?
Run "/install code-explore" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Code Explore free?
Yes, Code Explore is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Code Explore support?
Code Explore is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Code Explore?
It is built and maintained by sf0799 (@sf0799); the current version is v0.1.0.
More Skills