agent-from-scratch — security grade SAFE, quality 64/100

Security audit verdict: SAFE · quality 64/100

No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →

by liiiiiiiiil · Codex Skill · ★ 91

Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h

🔒 Is agent-from-scratch safe to install? View the security audit →

About agent-from-scratch

agent-from-scratch 逐步生长的编程 Agent:从零开始构建一个能干活的 AI Agent 从最小的 agent loop 开始,按 Git tag 和能力阶段逐步加入工具、安全、上下文和可靠执行能力。 English · 中文 适合想用 Python 标准库理解 LLM agent 如何运行的开发者。每课聚焦一个版本相对上一版新增的核心概念,源码、diff 和设计取舍都可追溯。 当前状态:主线代码和教程到 (第 25 课:计划轨迹回放与验收)。教程以默认分支的 为准;运行某课时再切换该课声明的代码 tag。 快速入口:运行 · 学习路径 · 学习指南 · 完整手册 学习路径 版本主题简介 阶段一 · 理解 Agent Loop v0.01最简 agent loop建立最小对话循环,理解请求、回复与结束条件。 阶段二 · 工具与安全 v0.02第一个工具接入 calculate,跑通 function calling 基本协议。 v0.03文件读写工具

agent-from-scratchagent-tutorialagentic-aiai-agentai-coding-assistantclaude-codecodexcoding-agentharnessllm-agent

Quick Facts

Stars91
Forks5
LanguagePython
CategoryCodex Skill
LicenseMIT
Quality Score64.4746126471781/100
Last Updated2026-09-28
Created2026-08-26
Platformsclaude-code, codex, python
Est. Tokens~15k

Compatible Skills

These tools work well together with agent-from-scratch for enhanced workflows:

  • ToolRegistry — semantic(0.19)+complementary+same_lang+similar_pop+shared_platform (57%)

agent-from-scratch alternative? Top 6 similar tools

Looking for a agent-from-scratch alternative? If you're comparing agent-from-scratch with other codex skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.

  • palot by ItsWendell · ⭐ 190

    Just another desktop client for OpenCode 2. Manage projects, sessions, parallel agents, requests, and changes

  • golembot by 0xranx · ⭐ 322

    Any Agent × Any Provider × Anywhere. Connect Cursor, Claude Code, OpenCode, or Codex to Slack, Telegram, Disco

  • speq-skill by marconae · ⭐ 54

    A light-weight and straightforward system for spec-driven development with Claude Code or OpenAI Codex. Writte

  • hotplex by hrygo · ⭐ 50

    HotPlex — Unified access layer for AI Coding Agent.

  • learn-hermes-agent by longyunfeigu · ⭐ 207

    A 27-chapter hands-on tutorial for building an autonomous AI agent from zero in Python. Agent loop, tool syste

  • roampal-core by roampal-ai · ⭐ 51

    Outcome-based persistent memory MCP server for Claude Code and OpenCode. Good advice promoted, bad advice demo

More Codex Skill Tools

Explore other popular codex skill tools:

View all Codex Skill tools →

Popular Python Agent Tools

Frequently Asked Questions

What is agent-from-scratch?

agent-from-scratch is Build an AI Coding Agent from Scratch, Step by Step | 从零一步步搭建 AI 编程 Agent,边实现边理解 Agent 的工作原理. It is categorized as a Codex Skill with 91 GitHub stars.

What programming language is agent-from-scratch written in?

agent-from-scratch is primarily written in Python. It covers topics such as agent-from-scratch, agent-tutorial, agentic-ai.

How do I install or use agent-from-scratch?

You can find installation instructions and usage details in the agent-from-scratch GitHub repository at github.com/liiiiiiiiil/agent-from-scratch. The project has 91 stars and 5 forks, indicating an active community.

What license does agent-from-scratch use?

agent-from-scratch is released under the MIT license, making it free to use and modify according to the license terms.

What are the best alternatives to agent-from-scratch?

The top alternatives to agent-from-scratch on Agent Skills Hub include palot, golembot, speq-skill. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.

How this security grade is produced

Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.

The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.

Sources & who's responsible:

View on GitHub → Browse Codex Skill tools