← 返回 Skills 市场
jhauga

Quasi Coder

作者 John Haugabook · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
804
总下载
0
收藏
4
当前安装
1
版本数
在 OpenClaw 中安装
/install quasi-coder
功能描述
Expert 10x engineer skill for interpreting and implementing code from shorthand, quasi-code, and natural language descriptions. Use when collaborators provid...
安全使用建议
This skill appears internally consistent for converting shorthand into code, but it gives the agent wide latitude to rewrite or ‘fix’ collaborators’ descriptions. Before using: (1) prefer user-invoked use (don’t enable autonomous edits), (2) keep work under version control and review diffs the skill produces, (3) run tests and security scans on generated code, (4) avoid embedding secrets in shorthand, and (5) ask the agent to request clarifying questions rather than making large assumptions when intent is unclear.
功能分析
Type: OpenClaw Skill Name: quasi-coder Version: 1.0.0 The `SKILL.md` file explicitly instructs the AI agent to execute 'non-code actions' such as 'run commands, create files, fetch data' based on user-provided shorthand input. This creates a severe prompt injection vulnerability, allowing an attacker to potentially achieve arbitrary command execution (RCE), file system manipulation, or data exfiltration by crafting malicious shorthand. The skill also instructs the agent to 'compensate for erroneous descriptions' and prioritize the 'goal over the method,' further increasing the risk of unintended harmful actions.
能力评估
Purpose & Capability
The name/description (translate shorthand/quasi-code into production code) matches the SKILL.md content. The skill declares no binaries, env vars, or install steps — all consistent with an instruction-only code-generation helper.
Instruction Scope
SKILL.md gives the agent broad, explicit rules for interpreting and replacing shorthand (markers, '()=>', confidence thresholds for changing approaches). This is coherent for a quasi-coding role, but it also gives the agent substantial discretion to transform collaborators' input into production code; users should expect potentially large automated edits and verify changes. The instructions do not instruct reading unrelated system files, secrets, or contacting external endpoints.
Install Mechanism
No install spec and no code files — lowest-risk model: nothing is downloaded or written by an installer as part of skill installation.
Credentials
The skill requires no environment variables, credentials, or config paths. There is no disproportionate request for secrets or external service tokens.
Persistence & Privilege
always:false and default autonomous invocation are set (normal). The skill does not request permanent presence or modifications to other skills or system-wide settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install quasi-coder
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /quasi-coder 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
# Version 1.0.0 Add initial `quasi-coder` skill that tasks the agent as a 10x software engineer who is working with a colleague or collaborator on a project or goal. The collaborator gives the skilled `quasi-coder` the big picture or general idea, and the skilled `quasi-coder` applies expert knowledge in order to achieve project goals.
元数据
Slug quasi-coder
版本 1.0.0
许可证
累计安装 4
当前安装数 4
历史版本数 1
常见问题

Quasi Coder 是什么?

Expert 10x engineer skill for interpreting and implementing code from shorthand, quasi-code, and natural language descriptions. Use when collaborators provid... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 804 次。

如何安装 Quasi Coder?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install quasi-coder」即可一键安装,无需额外配置。

Quasi Coder 是免费的吗?

是的,Quasi Coder 完全免费(开源免费),可自由下载、安装和使用。

Quasi Coder 支持哪些平台?

Quasi Coder 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Quasi Coder?

由 John Haugabook(@jhauga)开发并维护,当前版本 v1.0.0。

💬 留言讨论