ClawHub

Code Write

v0.1.0

Implement or modify code in the current project to match style and handle edge cases, focusing only on requested feature, bugfix, or test updates.

⭐ 0· 121·0 current·0 all-time
by@sf0799

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for sf0799/code-write.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Code Write" (sf0799/code-write) from ClawHub.
Skill page: https://clawhub.ai/sf0799/code-write
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install code-write

ClawHub CLI

Package manager switcher

npx clawhub@latest install code-write
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description match the runtime instructions: the skill is an in-repo code-writing helper. It requires no external binaries, credentials, or installs — appropriate for the stated purpose.
ℹ
Instruction Scope
SKILL.md explicitly directs the agent to read affected files, nearby callers, tests, and configs and to run minimal verification. This is appropriate for modifying code, but it implicitly grants the agent broad read access to the workspace; that may include sensitive config or credential files present in the repo. The instructions do not direct exfiltration or external network calls, nor do they request secrets, but users should be aware the agent will inspect repository files to produce edits.
✓
Install Mechanism
Instruction-only skill with no install spec or external downloads. Lowest-risk install surface.
✓
Credentials
No environment variables, credentials, or config paths are requested. The lack of requested secrets is proportional to a local code-editing task.
✓
Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent system presence or modify other skills' configuration. Autonomous invocation remains possible (platform default) but is not combined with other concerning privileges.
Assessment
This skill is coherent for editing code in your repository and does not ask for external credentials or installs. Before enabling it, consider: (1) the agent will read files in the workspace — remove or exclude any sensitive files (API keys, .env, private configs) you don't want inspected; (2) review changes (use version control or PRs) before accepting them; (3) running the skill's recommended verification (tests/build) may execute project code and could access network resources or secrets configured in the environment — run in an isolated sandbox if unsure; (4) prefer granting the agent access only to the repo or directory it truly needs. If you want stronger guarantees, require manual review of all edits or run the skill in a restricted environment.

Like a lobster shell, security has layers — review code before you run it.

latestvk973tcp6v74b4wvvjgwj5qxdx184angq
121downloads
0stars
1versions
Updated 3w ago
v0.1.0
MIT-0
@sf0799
latest
Download

代码编写

先读上下文，再只改需求要求的部分。

工作流

  1. 修改前先读受影响文件，以及附近的调用方、测试和配置。
  2. 严格匹配项目现有命名、格式、结构和异常处理风格。
  3. 只实现用户要求的行为，不顺手扩功能。
  4. 明确处理非法输入、空状态、资源缺失和失败路径。
  5. 不写死密钥、不伪造逻辑、不留下死代码和无用变量。
  6. 只有在代码难读时才补简短注释。
  7. 项目已有测试体系时，优先补或更新最相关的测试。
  8. 改完后运行最小必要验证。

约束

  • 没读过的文件不要直接改
  • 没要求的新功能不要自行添加
  • 除非项目本来就这样做，否则不要硬编码环境值
  • 优先最小且安全的差异

输出

  • 修改了哪些文件
  • 实现理由的简短说明
  • 已做验证，或未验证的阻塞原因

Comments

Loading comments...