mirror of
https://github.com/catlog22/Claude-Code-Workflow.git
synced 2026-02-12 02:37:45 +08:00
f14418603a
重构 ccw cli 模板系统： - 新增 template-discovery.ts 模块，支持扁平化模板自动发现 - 添加 --rule <template> 选项，自动加载 protocol 和 template - 模板目录从嵌套结构 (prompts/category/file.txt) 迁移到扁平结构 (prompts/category-function.txt) - 更新所有 agent/command 文件，使用 $PROTO $TMPL 环境变量替代 $(cat ...) 模式 - 支持模糊匹配：--rule 02-review-architecture 可匹配 analysis-review-architecture.txt 其他更新： - Dashboard: 添加 Claude Manager 和 Issue Manager 页面 - Codex-lens: 增强 chain_search 和 clustering 模块 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
90 lines
3.0 KiB
Plaintext
90 lines
3.0 KiB
Plaintext
# Tech Stack Rules Generation Agent Prompt
|
|
|
|
## Context Variables
|
|
- {TECH_STACK_NAME}: Normalized tech stack name (e.g., "typescript-react")
|
|
- {PRIMARY_LANG}: Primary language (e.g., "typescript")
|
|
- {FILE_EXT}: File extension pattern (e.g., "{ts,tsx}")
|
|
- {FRAMEWORK_TYPE}: frontend | backend | fullstack | library
|
|
- {COMPONENTS}: Array of tech components
|
|
- {OUTPUT_DIR}: .claude/rules/tech/{TECH_STACK_NAME}/
|
|
|
|
## Agent Instructions
|
|
|
|
Generate path-conditional rules for Claude Code automatic loading.
|
|
|
|
### Step 1: Execute Exa Research
|
|
|
|
Run 4-6 parallel queries based on tech stack:
|
|
|
|
**Base Queries** (always execute):
|
|
```
|
|
mcp__exa__get_code_context_exa(query: "{PRIMARY_LANG} best practices principles 2025", tokensNum: 8000)
|
|
mcp__exa__get_code_context_exa(query: "{PRIMARY_LANG} implementation patterns examples", tokensNum: 7000)
|
|
mcp__exa__get_code_context_exa(query: "{PRIMARY_LANG} testing strategies conventions", tokensNum: 5000)
|
|
mcp__exa__web_search_exa(query: "{PRIMARY_LANG} configuration setup 2025", numResults: 5)
|
|
```
|
|
|
|
**Component Queries** (for each framework in COMPONENTS):
|
|
```
|
|
mcp__exa__get_code_context_exa(query: "{PRIMARY_LANG} {component} integration patterns", tokensNum: 5000)
|
|
```
|
|
|
|
### Step 2: Read Rule Templates
|
|
|
|
Read each template file before generating content:
|
|
```
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-core.txt)
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-patterns.txt)
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-testing.txt)
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-config.txt)
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-api.txt) # Only if backend/fullstack
|
|
Read(~/.claude/workflows/cli-templates/prompts/rules/rule-components.txt) # Only if frontend/fullstack
|
|
```
|
|
|
|
### Step 3: Generate Rule Files
|
|
|
|
Create directory and write files:
|
|
```bash
|
|
mkdir -p "{OUTPUT_DIR}"
|
|
```
|
|
|
|
**Always Generate**:
|
|
- core.md (from rule-core.txt template)
|
|
- patterns.md (from rule-patterns.txt template)
|
|
- testing.md (from rule-testing.txt template)
|
|
- config.md (from rule-config.txt template)
|
|
|
|
**Conditional**:
|
|
- api.md: Only if FRAMEWORK_TYPE == 'backend' or 'fullstack'
|
|
- components.md: Only if FRAMEWORK_TYPE == 'frontend' or 'fullstack'
|
|
|
|
### Step 4: Write Metadata
|
|
|
|
```json
|
|
{
|
|
"tech_stack": "{TECH_STACK_NAME}",
|
|
"primary_lang": "{PRIMARY_LANG}",
|
|
"file_ext": "{FILE_EXT}",
|
|
"framework_type": "{FRAMEWORK_TYPE}",
|
|
"components": ["{COMPONENTS}"],
|
|
"generated_at": "{ISO_TIMESTAMP}",
|
|
"source": "exa-research",
|
|
"files_generated": ["core.md", "patterns.md", "testing.md", "config.md", ...]
|
|
}
|
|
```
|
|
|
|
### Step 5: Report Completion
|
|
|
|
Provide summary:
|
|
- Files created with their path patterns
|
|
- Exa queries executed (count)
|
|
- Sources consulted (count)
|
|
|
|
## Critical Requirements
|
|
|
|
1. Every .md file MUST start with `paths` YAML frontmatter
|
|
2. Use {FILE_EXT} consistently across all rule files
|
|
3. Synthesize Exa research into actionable rules
|
|
4. Include code examples from Exa sources
|
|
5. Keep each file focused on its specific domain
|