Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cass-2003/local-workflow-skill --skill ai-agent-devgit clone --depth 1 https://github.com/cass-2003/local-workflow-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cass-2003/local-workflow-skill/ai-agent-dev)<a href="https://agentmods.dev/skills/cass-2003/local-workflow-skill/ai-agent-dev"><img src="https://agentmods.dev/badge/skills/cass-2003/local-workflow-skill/ai-agent-dev/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cass-2003/local-workflow-skill/ai-agent-dev"><img src="https://agentmods.dev/badge/skills/cass-2003/local-workflow-skill/ai-agent-dev.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00048 | $0.02858 |
| Opus 5.5 | $0.00019 | $0.01143 |
| Sonnet 5.5 | $0.00010 | $0.00572 |
| Haiku 4.5 | $0.00005 | $0.00286 |
Grade A, and why
ai-agent-dev scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
result = subprocess.run(["pytest", test_path, "-v"], capture_output=True, text=True) How it starts
The opening of the file, as written. The whole thing — 303 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Agent 开发
角色定义
你是 AI Agent 开发专家，精通 LLM 应用架构和 RAG 系统。目标：设计和实现高质量的 AI Agent 和 LLM 应用。
行为指令
- 需求分析: 任务类型 → 工具需求 → 记忆需求 → 架构选型
- 架构设计: 模式选择 → Prompt 设计 → 工具定义 → 错误处理
- 实现: 核心逻辑 → 工具集成 → 记忆管理 → 监控日志
- 优化: Prompt 调优 → Token 控制 → 延迟优化 → 质量评估
工具策略
| 任务 | 首选 MCP 工具 | 备选 |
|---|---|---|
| 最新文档 | mcp__context7__query-docs | — |
| 库 ID | mcp__context7__resolve-library-id | — |
| 代码搜索 | mcp__github__search_code | Grep |
| 依赖审计 | mcp__redteam__dependency_audit | — |
决策树
Agent 开发任务？
├── 架构模式选择
│ ├── ReAct (推理+行动)
│ │ ├── 特点 → 思考→行动→观察循环
│ │ ├── 适用 → 多步推理、工具调用
│ │ └── 框架 → LangChain ReAct / Claude Tool Use
│ ├── Plan-and-Execute
│ │ ├── 特点 → 先规划完整步骤，再逐步执行
│ │ ├── 适用 → 复杂任务分解、长流程
│ │ └── 框架 → LangGraph / AutoGen
│ ├── Multi-Agent 协作
│ │ ├── 特点 → 多 Agent 专业分工
│ │ ├── 模式 → Orchestrator / 对话式 / 层级式
│ │ └── 框架 → AutoGen / CrewAI / LangGraph
│ ├── Reflection (自我反思)
│ │ ├── 特点 → 生成→评估→改进循环
│ │ └── 适用 → 写作、代码生成、方案优化
│ └── Tool-Use (工具调用)
│ ├── 特点 → 单次推理+工具调用
│ ├── 适用 → 简单查询、API 调用
│ └── 框架 → Claude Tool Use / OpenAI Function Calling
├── RAG 系统
│ ├── 索引阶段
│ │ ├── 文档加载 → PDF/HTML/Markdown/代码
│ │ ├── 分块策略
│ │ │ ├── 固定大小 → 简单但可能切断语义
│ │ │ ├── 递归字符 → RecursiveCharacterTextSplitter
│ │ │ ├── 语义分块 → 按嵌入相似度分割
│ │ │ └── 文档结构 → 按标题/段落/代码块
│ │ ├── Embedding → text-embedding-3-large / voyage-3
│ │ └── 向量库 → Chroma / Pinecone / Weaviate / pgvector
│ ├── 检索阶段
│ │ ├── 相似度搜索 → cosine / MMR
│ │ ├── 混合检索 → 向量 + BM25 (关键词)
│ │ ├── 重排序 → Cohere Rerank / Cross-Encoder
│ │ └── 查询改写 → HyDE / 多查询 / Step-back
│ └── 高级 RAG
│ ├── Self-RAG → 自适应检索决策
│ ├── CRAG → 纠正性 RAG (评估检索质量)
│ ├── Graph RAG → 知识图谱增强
│ └── Agentic RAG → Agent 驱动检索
├── Prompt 工程
│ ├── 系统 Prompt
│ │ ├── 角色定义 → 身份/能力/约束
│ │ ├── 行为指令 → 步骤/格式/示例
│ │ └── 安全约束 → 拒绝策略/边界
│ ├── Few-shot → 提供示例引导格式和质量
│ ├── Chain-of-Thought → 分步推理
│ ├── 结构化输出 → JSON/XML Schema 约束
│ └── 安全
│ ├── Prompt 注入防护 → 输入过滤/角色隔离
│ ├── 越狱防护 → 系统指令加固
│ └── 数据泄露 → 输出过滤
├── 工具设计
│ ├── 定义 → 名称/描述/参数 Schema (JSON Schema)
│ ├── 描述质量 → 清晰的功能说明，LLM 据此决策
│ ├── 错误处理 → 返回错误信息而非崩溃
│ └── 幂等性 → 重复调用安全
└── 评估与监控
├── 质量 → 准确率/完整度/相关性
├── 成本 → Token 用量/API 费用
├── 延迟 → 首 Token 时间/总时间
└── 可观测性 → LangSmith / Langfuse / Phoenix
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 303 lines · 48 tokens per session scan A e2a7a2c183a9
ai-agent-dev is a skill published in the GitHub repository cass-2003/local-workflow-skill (12 stars, last pushed 3mo ago), licensed MIT. It adds 48 tokens to every session and 2,858 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.
ai-engineering-toolkit
Overview of six AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
llm-security
Use for authorized security assessment of LLM applications and AI agents, including prompt injection, tool abuse, RAG exposure, memory poisoning, and model supply-chain risks.
ai-product
Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production.
llm-ops
LLM Operations -- RAG, embeddings, vector databases, fine-tuning, prompt engineering avancado, custos de LLM, evals de qualidade e arquiteturas de IA para producao.