agent-configs 1.0.0 → 1.0.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +12 -8
- package/bundles/docs-bundle/.claude/commands/update-codemaps.md +17 -0
- package/bundles/docs-bundle/.claude/commands/update-docs.md +31 -0
- package/bundles/docs-bundle/.claude/settings.json +29 -0
- package/bundles/docs-bundle/.cursor/commands/update-codemaps.md +17 -0
- package/bundles/docs-bundle/.cursor/commands/update-docs.md +31 -0
- package/bundles/docs-bundle/.cursor/hooks.json +19 -0
- package/bundles/review-bundle/.claude/agents/code-reviewer.md +104 -0
- package/bundles/review-bundle/.claude/agents/security-reviewer.md +545 -0
- package/bundles/review-bundle/.claude/commands/code-review.md +40 -0
- package/{hooks/hooks.json → bundles/review-bundle/.claude/settings.json} +4 -34
- package/bundles/review-bundle/.claude/skills/security-review/SKILL.md +497 -0
- package/bundles/review-bundle/.claude/skills/security-review/references/cloud-infrastructure-security.md +361 -0
- package/bundles/review-bundle/.cursor/agents/code-reviewer.md +104 -0
- package/bundles/review-bundle/.cursor/agents/security-reviewer.md +545 -0
- package/bundles/review-bundle/.cursor/commands/code-review.md +40 -0
- package/bundles/review-bundle/.cursor/hooks.json +19 -0
- package/bundles/review-bundle/.cursor/skills/security-review/SKILL.md +497 -0
- package/bundles/review-bundle/.cursor/skills/security-review/references/cloud-infrastructure-security.md +361 -0
- package/cli.js +10 -8
- package/configs.json +10 -20
- package/package.json +1 -1
- package/skills/ai-config-architect/references/hooks.md +122 -19
package/README.md
CHANGED
|
@@ -8,7 +8,7 @@ CLI tool to install curated Claude Code and Cursor configurations.
|
|
|
8
8
|
|
|
9
9
|
- **交互式菜单** - 上下选择、Enter 进入/安装、Esc 返回
|
|
10
10
|
- **双平台支持** - Claude Code + Cursor(默认 Cursor)
|
|
11
|
-
- **完整配置** - Commands + Skills + Rules + Agents + Hooks
|
|
11
|
+
- **完整配置** - Commands + Skills + Rules + Agents + Bundles(Hooks 通过 Bundles 提供)
|
|
12
12
|
- **Bundle 优先** - 相关配置打包,一键安装
|
|
13
13
|
|
|
14
14
|
## 快速开始
|
|
@@ -41,11 +41,10 @@ agent-configs - Claude Code & Cursor 配置共享工具
|
|
|
41
41
|
|
|
42
42
|
? 选择类别
|
|
43
43
|
❯ Commands (10) ★2 (claude/cursor)
|
|
44
|
-
Skills (
|
|
44
|
+
Skills (8) (claude/cursor)
|
|
45
45
|
Rules (8) ★1 (cursor)
|
|
46
46
|
Agents (6) ★2 (claude/cursor)
|
|
47
|
-
|
|
48
|
-
Bundles (5) ★3
|
|
47
|
+
Bundles (6) ★4
|
|
49
48
|
──────────────
|
|
50
49
|
退出
|
|
51
50
|
```
|
|
@@ -71,10 +70,11 @@ agent-configs - Claude Code & Cursor 配置共享工具
|
|
|
71
70
|
| update-docs | 同步文档 |
|
|
72
71
|
| update-codemaps | 更新架构文档 |
|
|
73
72
|
|
|
74
|
-
### Skills(
|
|
73
|
+
### Skills(8 个)
|
|
75
74
|
|
|
76
75
|
| Skill | 说明 |
|
|
77
76
|
|-------|------|
|
|
77
|
+
| ai-config-architect | AI 配置架构师 - 生成 rules/skills/commands/hooks/agents 配置,支持 Claude/Cursor 双平台 |
|
|
78
78
|
| tdd-workflow | TDD 完整工作流和最佳实践 |
|
|
79
79
|
| security-review | 安全审查检查清单 |
|
|
80
80
|
| coding-standards | 编码规范和风格指南 |
|
|
@@ -94,7 +94,7 @@ agent-configs - Claude Code & Cursor 配置共享工具
|
|
|
94
94
|
| agents | 子代理使用规范 |
|
|
95
95
|
| patterns | API 响应格式规范 |
|
|
96
96
|
| performance | 模型选择和上下文管理 |
|
|
97
|
-
| hooks | Hook 使用文档 |
|
|
97
|
+
| hooks-guide | Hook 使用文档 |
|
|
98
98
|
|
|
99
99
|
### Agents(6 个)
|
|
100
100
|
|
|
@@ -107,15 +107,17 @@ agent-configs - Claude Code & Cursor 配置共享工具
|
|
|
107
107
|
| security-reviewer | 安全审查专家 |
|
|
108
108
|
| refactor-cleaner | 重构清理专家 |
|
|
109
109
|
|
|
110
|
-
### Bundles(
|
|
110
|
+
### Bundles(6 个)
|
|
111
111
|
|
|
112
112
|
| Bundle | 包含内容 | 说明 |
|
|
113
113
|
|--------|----------|------|
|
|
114
114
|
| shared-memory | commands + hooks + lib + skills + rules | 跨会话记忆管理系统 |
|
|
115
115
|
| tdd-bundle | tdd + tdd-workflow + tdd-guide | TDD 完整工作流 |
|
|
116
|
-
| review-bundle | code-review + security-review + code-reviewer + security-reviewer | 代码审查全套 |
|
|
116
|
+
| review-bundle | code-review + security-review + code-reviewer + security-reviewer + hooks | 代码审查全套 - 包含 /code-review 命令、审查 agents、Prettier/TypeScript/console.log 检查 hooks |
|
|
117
117
|
| planning-bundle | plan + planner + architect | 规划和架构设计 |
|
|
118
118
|
| bk-chat-bundle | bk-chat-helper + bk-chat-x | 蓝鲸 AI Chat 开发 |
|
|
119
|
+
| continuous-learning | commands + hooks + lib + skills + rules | 持续学习 - 从会话中自动提取 patterns,保存为 skills/instincts |
|
|
120
|
+
| docs-bundle | commands + hooks | 文档自动化 - 包含 /update-docs、/update-codemaps 命令及文档更新提醒 hooks |
|
|
119
121
|
|
|
120
122
|
## 使用方式
|
|
121
123
|
|
|
@@ -158,6 +160,8 @@ npx agent-configs install review-bundle
|
|
|
158
160
|
npx agent-configs install planning-bundle
|
|
159
161
|
npx agent-configs install bk-chat-bundle
|
|
160
162
|
npx agent-configs install shared-memory
|
|
163
|
+
npx agent-configs install continuous-learning
|
|
164
|
+
npx agent-configs install docs-bundle
|
|
161
165
|
|
|
162
166
|
# 指定平台
|
|
163
167
|
npx agent-configs install plan --agent claude
|
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
# Update Codemaps
|
|
2
|
+
|
|
3
|
+
Analyze the codebase structure and update architecture documentation:
|
|
4
|
+
|
|
5
|
+
1. Scan all source files for imports, exports, and dependencies
|
|
6
|
+
2. Generate token-lean codemaps in the following format:
|
|
7
|
+
- codemaps/architecture.md - Overall architecture
|
|
8
|
+
- codemaps/backend.md - Backend structure
|
|
9
|
+
- codemaps/frontend.md - Frontend structure
|
|
10
|
+
- codemaps/data.md - Data models and schemas
|
|
11
|
+
|
|
12
|
+
3. Calculate diff percentage from previous version
|
|
13
|
+
4. If changes > 30%, request user approval before updating
|
|
14
|
+
5. Add freshness timestamp to each codemap
|
|
15
|
+
6. Save reports to .reports/codemap-diff.txt
|
|
16
|
+
|
|
17
|
+
Use TypeScript/Node.js for analysis. Focus on high-level structure, not implementation details.
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
# Update Documentation
|
|
2
|
+
|
|
3
|
+
Sync documentation from source-of-truth:
|
|
4
|
+
|
|
5
|
+
1. Read package.json scripts section
|
|
6
|
+
- Generate scripts reference table
|
|
7
|
+
- Include descriptions from comments
|
|
8
|
+
|
|
9
|
+
2. Read .env.example
|
|
10
|
+
- Extract all environment variables
|
|
11
|
+
- Document purpose and format
|
|
12
|
+
|
|
13
|
+
3. Generate docs/CONTRIB.md with:
|
|
14
|
+
- Development workflow
|
|
15
|
+
- Available scripts
|
|
16
|
+
- Environment setup
|
|
17
|
+
- Testing procedures
|
|
18
|
+
|
|
19
|
+
4. Generate docs/RUNBOOK.md with:
|
|
20
|
+
- Deployment procedures
|
|
21
|
+
- Monitoring and alerts
|
|
22
|
+
- Common issues and fixes
|
|
23
|
+
- Rollback procedures
|
|
24
|
+
|
|
25
|
+
5. Identify obsolete documentation:
|
|
26
|
+
- Find docs not modified in 90+ days
|
|
27
|
+
- List for manual review
|
|
28
|
+
|
|
29
|
+
6. Show diff summary
|
|
30
|
+
|
|
31
|
+
Single source of truth: package.json and .env.example
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json.schemastore.org/claude-code-settings.json",
|
|
3
|
+
"hooks": {
|
|
4
|
+
"PreToolUse": [
|
|
5
|
+
{
|
|
6
|
+
"matcher": "tool == \"Write\" && tool_input.file_path matches \"\\\\.(md|txt)$\" && !(tool_input.file_path matches \"README\\\\.md|CLAUDE\\\\.md|AGENTS\\\\.md|CONTRIBUTING\\\\.md|CHANGELOG\\\\.md\")",
|
|
7
|
+
"hooks": [
|
|
8
|
+
{
|
|
9
|
+
"type": "command",
|
|
10
|
+
"command": "node -e \"const fs=require('fs');let d='';process.stdin.on('data',c=>d+=c);process.stdin.on('end',()=>{const i=JSON.parse(d);const p=i.tool_input?.file_path||'';if(/\\.(md|txt)$/.test(p)&&!/(README|CLAUDE|AGENTS|CONTRIBUTING|CHANGELOG)\\.md$/.test(p)&&!/codemaps\\//.test(p)&&!/docs\\//.test(p)){console.error('[Hook] BLOCKED: Unnecessary documentation file creation');console.error('[Hook] File: '+p);console.error('[Hook] Use README.md or docs/ directory instead');process.exit(1)}console.log(d)})\""
|
|
11
|
+
}
|
|
12
|
+
],
|
|
13
|
+
"description": "Block creation of random .md files - keeps docs consolidated"
|
|
14
|
+
}
|
|
15
|
+
],
|
|
16
|
+
"Stop": [
|
|
17
|
+
{
|
|
18
|
+
"matcher": "*",
|
|
19
|
+
"hooks": [
|
|
20
|
+
{
|
|
21
|
+
"type": "command",
|
|
22
|
+
"command": "node -e \"const{execSync}=require('child_process');const fs=require('fs');try{const diff=execSync('git diff --name-only HEAD 2>/dev/null',{encoding:'utf8'});const staged=execSync('git diff --name-only --cached 2>/dev/null',{encoding:'utf8'});const all=[...diff.split('\\n'),...staged.split('\\n')].filter(Boolean);const code=all.filter(f=>/\\.(js|ts|tsx|jsx|json)$/.test(f)&&!/package-lock|pnpm-lock/.test(f));const docs=all.filter(f=>/README\\.md|CLAUDE\\.md|CHANGELOG/.test(f));if(code.length>0&&docs.length===0){console.log(JSON.stringify({reminder:'📝 Code changed but docs not updated. Modified: '+code.slice(0,3).join(', ')+(code.length>3?' (+'+(code.length-3)+' more)':'')+'. Consider: /update-docs'}));}}catch(e){}\""
|
|
23
|
+
}
|
|
24
|
+
],
|
|
25
|
+
"description": "Remind to update docs when code changes but README unchanged"
|
|
26
|
+
}
|
|
27
|
+
]
|
|
28
|
+
}
|
|
29
|
+
}
|
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
# Update Codemaps
|
|
2
|
+
|
|
3
|
+
Analyze the codebase structure and update architecture documentation:
|
|
4
|
+
|
|
5
|
+
1. Scan all source files for imports, exports, and dependencies
|
|
6
|
+
2. Generate token-lean codemaps in the following format:
|
|
7
|
+
- codemaps/architecture.md - Overall architecture
|
|
8
|
+
- codemaps/backend.md - Backend structure
|
|
9
|
+
- codemaps/frontend.md - Frontend structure
|
|
10
|
+
- codemaps/data.md - Data models and schemas
|
|
11
|
+
|
|
12
|
+
3. Calculate diff percentage from previous version
|
|
13
|
+
4. If changes > 30%, request user approval before updating
|
|
14
|
+
5. Add freshness timestamp to each codemap
|
|
15
|
+
6. Save reports to .reports/codemap-diff.txt
|
|
16
|
+
|
|
17
|
+
Use TypeScript/Node.js for analysis. Focus on high-level structure, not implementation details.
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
# Update Documentation
|
|
2
|
+
|
|
3
|
+
Sync documentation from source-of-truth:
|
|
4
|
+
|
|
5
|
+
1. Read package.json scripts section
|
|
6
|
+
- Generate scripts reference table
|
|
7
|
+
- Include descriptions from comments
|
|
8
|
+
|
|
9
|
+
2. Read .env.example
|
|
10
|
+
- Extract all environment variables
|
|
11
|
+
- Document purpose and format
|
|
12
|
+
|
|
13
|
+
3. Generate docs/CONTRIB.md with:
|
|
14
|
+
- Development workflow
|
|
15
|
+
- Available scripts
|
|
16
|
+
- Environment setup
|
|
17
|
+
- Testing procedures
|
|
18
|
+
|
|
19
|
+
4. Generate docs/RUNBOOK.md with:
|
|
20
|
+
- Deployment procedures
|
|
21
|
+
- Monitoring and alerts
|
|
22
|
+
- Common issues and fixes
|
|
23
|
+
- Rollback procedures
|
|
24
|
+
|
|
25
|
+
5. Identify obsolete documentation:
|
|
26
|
+
- Find docs not modified in 90+ days
|
|
27
|
+
- List for manual review
|
|
28
|
+
|
|
29
|
+
6. Show diff summary
|
|
30
|
+
|
|
31
|
+
Single source of truth: package.json and .env.example
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
{
|
|
2
|
+
"version": 1,
|
|
3
|
+
"hooks": {
|
|
4
|
+
"preToolUse": [
|
|
5
|
+
{
|
|
6
|
+
"matcher": "Write",
|
|
7
|
+
"command": "node -e \"const fs=require('fs');let d='';process.stdin.on('data',c=>d+=c);process.stdin.on('end',()=>{const i=JSON.parse(d);const p=i.tool_input?.file_path||'';if(/\\.(md|txt)$/.test(p)&&!/(README|CLAUDE|AGENTS|CONTRIBUTING|CHANGELOG)\\.md$/.test(p)&&!/codemaps\\//.test(p)&&!/docs\\//.test(p)){console.error('[Hook] BLOCKED: Unnecessary documentation file creation');console.error('[Hook] File: '+p);console.error('[Hook] Use README.md or docs/ directory instead');process.exit(2)}})\"",
|
|
8
|
+
"description": "Block creation of random .md files - keeps docs consolidated"
|
|
9
|
+
}
|
|
10
|
+
],
|
|
11
|
+
"stop": [
|
|
12
|
+
{
|
|
13
|
+
"command": "node -e \"const{execSync}=require('child_process');try{const diff=execSync('git diff --name-only HEAD 2>/dev/null',{encoding:'utf8'});const staged=execSync('git diff --name-only --cached 2>/dev/null',{encoding:'utf8'});const all=[...diff.split('\\n'),...staged.split('\\n')].filter(Boolean);const code=all.filter(f=>/\\.(js|ts|tsx|jsx|json)$/.test(f)&&!/package-lock|pnpm-lock/.test(f));const docs=all.filter(f=>/README\\.md|CLAUDE\\.md|CHANGELOG/.test(f));if(code.length>0&&docs.length===0){const files=code.slice(0,3).join(', ')+(code.length>3?' (+'+(code.length-3)+' more)':'');console.log(JSON.stringify({followup_message:'[文档提醒] 代码已修改但文档未更新。修改的文件: '+files+'。如需更新文档请使用 /update-docs,否则请忽略此消息。'}));}else{console.log('{}');}}catch(e){console.log('{}');}\"",
|
|
14
|
+
"loop_limit": 1,
|
|
15
|
+
"description": "Remind to update docs when code changes but README unchanged"
|
|
16
|
+
}
|
|
17
|
+
]
|
|
18
|
+
}
|
|
19
|
+
}
|
|
@@ -0,0 +1,104 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: code-reviewer
|
|
3
|
+
description: Expert code review specialist. Proactively reviews code for quality, security, and maintainability. Use immediately after writing or modifying code. MUST BE USED for all code changes.
|
|
4
|
+
tools: ["Read", "Grep", "Glob", "Bash"]
|
|
5
|
+
model: opus
|
|
6
|
+
---
|
|
7
|
+
|
|
8
|
+
You are a senior code reviewer ensuring high standards of code quality and security.
|
|
9
|
+
|
|
10
|
+
When invoked:
|
|
11
|
+
1. Run git diff to see recent changes
|
|
12
|
+
2. Focus on modified files
|
|
13
|
+
3. Begin review immediately
|
|
14
|
+
|
|
15
|
+
Review checklist:
|
|
16
|
+
- Code is simple and readable
|
|
17
|
+
- Functions and variables are well-named
|
|
18
|
+
- No duplicated code
|
|
19
|
+
- Proper error handling
|
|
20
|
+
- No exposed secrets or API keys
|
|
21
|
+
- Input validation implemented
|
|
22
|
+
- Good test coverage
|
|
23
|
+
- Performance considerations addressed
|
|
24
|
+
- Time complexity of algorithms analyzed
|
|
25
|
+
- Licenses of integrated libraries checked
|
|
26
|
+
|
|
27
|
+
Provide feedback organized by priority:
|
|
28
|
+
- Critical issues (must fix)
|
|
29
|
+
- Warnings (should fix)
|
|
30
|
+
- Suggestions (consider improving)
|
|
31
|
+
|
|
32
|
+
Include specific examples of how to fix issues.
|
|
33
|
+
|
|
34
|
+
## Security Checks (CRITICAL)
|
|
35
|
+
|
|
36
|
+
- Hardcoded credentials (API keys, passwords, tokens)
|
|
37
|
+
- SQL injection risks (string concatenation in queries)
|
|
38
|
+
- XSS vulnerabilities (unescaped user input)
|
|
39
|
+
- Missing input validation
|
|
40
|
+
- Insecure dependencies (outdated, vulnerable)
|
|
41
|
+
- Path traversal risks (user-controlled file paths)
|
|
42
|
+
- CSRF vulnerabilities
|
|
43
|
+
- Authentication bypasses
|
|
44
|
+
|
|
45
|
+
## Code Quality (HIGH)
|
|
46
|
+
|
|
47
|
+
- Large functions (>50 lines)
|
|
48
|
+
- Large files (>800 lines)
|
|
49
|
+
- Deep nesting (>4 levels)
|
|
50
|
+
- Missing error handling (try/catch)
|
|
51
|
+
- console.log statements
|
|
52
|
+
- Mutation patterns
|
|
53
|
+
- Missing tests for new code
|
|
54
|
+
|
|
55
|
+
## Performance (MEDIUM)
|
|
56
|
+
|
|
57
|
+
- Inefficient algorithms (O(n²) when O(n log n) possible)
|
|
58
|
+
- Unnecessary re-renders in React
|
|
59
|
+
- Missing memoization
|
|
60
|
+
- Large bundle sizes
|
|
61
|
+
- Unoptimized images
|
|
62
|
+
- Missing caching
|
|
63
|
+
- N+1 queries
|
|
64
|
+
|
|
65
|
+
## Best Practices (MEDIUM)
|
|
66
|
+
|
|
67
|
+
- Emoji usage in code/comments
|
|
68
|
+
- TODO/FIXME without tickets
|
|
69
|
+
- Missing JSDoc for public APIs
|
|
70
|
+
- Accessibility issues (missing ARIA labels, poor contrast)
|
|
71
|
+
- Poor variable naming (x, tmp, data)
|
|
72
|
+
- Magic numbers without explanation
|
|
73
|
+
- Inconsistent formatting
|
|
74
|
+
|
|
75
|
+
## Review Output Format
|
|
76
|
+
|
|
77
|
+
For each issue:
|
|
78
|
+
```
|
|
79
|
+
[CRITICAL] Hardcoded API key
|
|
80
|
+
File: src/api/client.ts:42
|
|
81
|
+
Issue: API key exposed in source code
|
|
82
|
+
Fix: Move to environment variable
|
|
83
|
+
|
|
84
|
+
const apiKey = "sk-abc123"; // ❌ Bad
|
|
85
|
+
const apiKey = process.env.API_KEY; // ✓ Good
|
|
86
|
+
```
|
|
87
|
+
|
|
88
|
+
## Approval Criteria
|
|
89
|
+
|
|
90
|
+
- ✅ Approve: No CRITICAL or HIGH issues
|
|
91
|
+
- ⚠️ Warning: MEDIUM issues only (can merge with caution)
|
|
92
|
+
- ❌ Block: CRITICAL or HIGH issues found
|
|
93
|
+
|
|
94
|
+
## Project-Specific Guidelines (Example)
|
|
95
|
+
|
|
96
|
+
Add your project-specific checks here. Examples:
|
|
97
|
+
- Follow MANY SMALL FILES principle (200-400 lines typical)
|
|
98
|
+
- No emojis in codebase
|
|
99
|
+
- Use immutability patterns (spread operator)
|
|
100
|
+
- Verify database RLS policies
|
|
101
|
+
- Check AI integration error handling
|
|
102
|
+
- Validate cache fallback behavior
|
|
103
|
+
|
|
104
|
+
Customize based on your project's `CLAUDE.md` or skill files.
|