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.
git clone --depth 1 https://github.com/clxzl/claude-code-best-practice-cnWrote 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/commands/clxzl/claude-code-best-practice-cn/research)<a href="https://agentmods.dev/commands/clxzl/claude-code-best-practice-cn/research"><img src="https://agentmods.dev/badge/commands/clxzl/claude-code-best-practice-cn/research.svg" alt="Measured on agentmods" 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 | $0.00018 | $0.03257 |
| Opus 5 | $0.00009 | $0.01629 |
| Sonnet 5 | $0.00004 | $0.00651 |
| Haiku 4.5 | $0.00002 | $0.00326 |
Grade A, and why
research scanned grade A with 0 findings 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 7d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.
用户输入
$ARGUMENTS
你必须解析用户输入以提取功能标识(rpi/ 中的文件夹名称)。
期望输入格式:rpi/{feature-slug}/REQUEST.md
目的
此命令在规划阶段开始之前对功能请求进行全面研究和分析。它作为一个关键的 GO/NO-GO 门禁,决定一个功能想法是否应进入详细规划。
关键目标:
- 评估产品市场匹配度和用户价值
- 评估技术可行性和复杂性
- 识别风险和潜在障碍
- 确定正确的方法(构建、购买、合作或拒绝)
- 做出 go/no-go 建议并给出清晰理由
前提条件:
- 功能文件夹存在于
rpi/{feature-slug}/ - 功能请求文件存在于
rpi/{feature-slug}/REQUEST.md
输出位置:rpi/{feature-slug}/research/RESEARCH.md
这是 RPI 工作流的第 2 步(在第 1 步初始功能描述之后)。
大纲
- 加载上下文:读取
rpi/{feature-slug}/中的功能描述和项目章程(如存在) - 解析功能请求:使用 requirement-parser Agent 提取结构化需求
- 执行多阶段研究:
- 阶段 1:解析功能请求(requirement-parser Agent)
- 阶段 2:产品分析与章程对齐(product-manager Agent)
- 阶段 2.5:技术探索(Explore Agent)- 关键:深度代码探索
- 阶段 3:技术可行性(senior-software-engineer Agent)
- 阶段 4:战略评估(technical-cto-advisor Agent)
- 阶段 5:生成研究报告(documentation-analyst-writer Agent)
- 综合建议:将所有分析合并为清晰的 go/no-go 建议
- 验证输出:根据质量门禁检查
- 报告完成:提供建议、后续步骤和报告位置
阶段
阶段 0:加载上下文
前提条件:已提供功能标识,rpi/{feature-slug}/REQUEST.md 存在
流程:
-
读取功能描述:
- 读取
rpi/{feature-slug}/REQUEST.md(必需) - 从 REQUEST.md 提取功能需求和目标
- 读取
-
检查项目章程(可选):
- 在仓库中查找章程或原则文档
- 常见位置:
constitution.md、PRINCIPLES.md、.project/constitution.md - 如果找到,提取核心原则、约束和目标
-
创建研究上下文:
- 为 Agent 合成简洁摘要
- 识别关键对齐标准
输出:
- 功能描述摘要
- 章程原则(如找到)
- 用于评估的对齐标准
验证:
- 功能文件夹存在于
rpi/{feature-slug}/ - 功能描述已提取
- 章程已检查并加载(如存在)
阶段 1:解析功能请求
前提条件:阶段 0 完成
Agent:requirement-parser(规划领域)
流程:
-
启动 requirement-parser Agent,传入功能描述
-
Agent 提取:
- 功能名称和类型
- 目标组件
- 目标和目的
- 功能性和非功能性需求
- 约束和假设
- 复杂度估计
- 澄清问题(如有)
-
审查解析结果:
- 如果存在澄清问题,停止并向用户提问后再继续
输出:
- 结构化需求文档
- 功能元数据(名称、类型、组件、复杂度)
- 澄清问题(如有)
阶段 2:产品分析与章程对齐
前提条件:阶段 1 完成,需求明确
Agent:product-manager
流程:
-
启动 product-manager Agent,传入:
- 阶段 1 的解析需求
- 阶段 0 的章程上下文
-
Agent 分析:
- 用户价值:谁受益?影响有多大?
- 市场匹配度:这是否符合市场需求?
- 产品愿景:这是否符合我们的产品策略?
- 章程对齐:这是否与项目原则一致?
- 约束检查:这是否违反任何章程约束?
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.
- 7d ago First seen · 382 lines · 18 tokens per session scan A 4693e126186f
research is a command published in the GitHub repository clxzl/claude-code-best-practice-cn (127 stars, last pushed 3mo ago), licensed MIT. It adds 18 tokens to every session and 3,257 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.