context-discovery
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseContext Discovery
Context Discovery
Overview
概述
Discover project standards and patterns before implementing features. Context-scout resolves the context root using the OAC Context Discovery Protocol, then finds and ranks relevant files based on your request.
Announce at start: "I'm using the context-discovery skill to find relevant standards for [feature/task]."
在实现功能前先了解项目标准和模式。context-scout会通过OAC Context Discovery Protocol解析上下文根目录,然后根据你的请求查找并排序相关文件。
开始时需告知:“我正在使用context-discovery技能来查找[功能/任务]的相关标准。”
The Process
流程
Step 1: Invoke Context-Scout
步骤1:调用Context-Scout
Run the skill with your implementation topic:
bash
/context-discovery [what you're implementing]Examples:
/context-discovery JWT authentication/context-discovery React form validation/context-discovery database migration workflow
针对你的实现主题运行该技能:
bash
/context-discovery [你要实现的内容]示例:
/context-discovery JWT authentication/context-discovery React form validation/context-discovery database migration workflow
Step 2: Load Critical Priority Files
步骤2:加载最高优先级文件
Read EVERY file marked Critical Priority (paths returned by context-scout, relative to the resolved context root):
bash
Read: {context_root}/core/standards/code-quality.md
Read: {context_root}/core/standards/security-patterns.mdThese are mandatory—proceed only after loading.
读取所有标记为Critical Priority(由context-scout返回的路径,相对于解析后的上下文根目录)的文件:
bash
Read: {context_root}/core/standards/code-quality.md
Read: {context_root}/core/standards/security-patterns.md这些文件是强制性的——必须加载完成后才能继续。
Step 3: Load High Priority Files
步骤3:加载高优先级文件
Read files marked High Priority:
bash
Read: {context_root}/core/workflows/approval-gates.mdThese are strongly recommended for your implementation.
读取标记为High Priority的文件:
bash
Read: {context_root}/core/workflows/approval-gates.md这些文件对于你的实现是强烈推荐的。
Step 4: Load Medium Priority (If Needed)
步骤4:加载中等优先级文件(如有需要)
Read Medium Priority files for additional context:
bash
Read: {context_root}/project-intelligence/architecture.mdThese are optional but helpful.
读取Medium Priority文件以获取额外上下文:
bash
Read: {context_root}/project-intelligence/architecture.md这些文件可选但有帮助。
Step 5: Apply to Implementation
步骤5:应用到实现中
- Follow standards from loaded files
- Apply patterns to your code
- Use naming conventions discovered
- Check workflows before executing
- 遵循已加载文件中的标准
- 将模式应用到你的代码中
- 使用发现的命名约定
- 执行前检查工作流程
Delegation Pattern
委托模式
When invoking subagents, pass discovered context files:
markdown
Invoke coder-agent:
Task: Implement JWT service
Context to load:
- .opencode/context/core/standards/code-quality.md
- .opencode/context/core/standards/security-patterns.md
Instructions: Follow functional patterns and security best practices.调用子Agent时,传递已发现的上下文文件:
markdown
Invoke coder-agent:
Task: Implement JWT service
Context to load:
- .opencode/context/core/standards/code-quality.md
- .opencode/context/core/standards/security-patterns.md
Instructions: Follow functional patterns and security best practices.Error Handling
错误处理
"No context files found"
- Run to download context first
/install-context
"Too many files returned"
- Be more specific (e.g., "TypeScript coding standards" not "coding")
"Which files do I load?"
- Always: Critical → High → Medium (if needed)
“未找到上下文文件”
- 先运行下载上下文
/install-context
“返回的文件过多”
- 请更具体(例如,使用“TypeScript coding standards”而非“coding”)
“我应该加载哪些文件?”
- 始终按照:最高优先级 → 高优先级 → 中等优先级(如有需要)的顺序
Remember
注意事项
- Context FIRST, code SECOND—never skip discovery
- Critical priority files are MANDATORY, not optional
- Training data is outdated—context is current
- Pass context forward when delegating to subagents
- Only use file paths returned by context-scout
- 先上下文,后代码——绝不要跳过发现步骤
- 最高优先级文件是强制性的,而非可选
- 训练数据已过时——上下文是最新的
- 委托给子Agent时传递上下文
- 仅使用context-scout返回的文件路径
Related
相关内容
- task-breakdown
- code-execution
- external-research
Task: Discover context files for $ARGUMENTS
Follow navigation-driven discovery and return ranked recommendations.
- task-breakdown
- code-execution
- external-research
任务:为**$ARGUMENTS**发现上下文文件
遵循导航式发现并返回排序后的推荐结果。