codingbuddy-rules 0.0.0-canary.20260106105956.dda0e31 → 0.0.0-canary.20260106121118.173ad8b
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/.ai-rules/adapters/cursor.md +120 -98
- package/package.json +1 -1
|
@@ -1,151 +1,173 @@
|
|
|
1
1
|
# Cursor Integration Guide
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
Guide for using codingbuddy with Cursor.
|
|
4
4
|
|
|
5
5
|
## Overview
|
|
6
6
|
|
|
7
|
-
|
|
7
|
+
codingbuddy integrates with Cursor in two ways:
|
|
8
8
|
|
|
9
|
-
|
|
9
|
+
1. **AGENTS.md** - Industry standard format compatible with all AI tools
|
|
10
|
+
2. **.cursor/rules/*.mdc** - Cursor-specific optimization (glob-based auto-activation)
|
|
10
11
|
|
|
11
|
-
|
|
12
|
+
## Two Usage Contexts
|
|
12
13
|
|
|
13
|
-
|
|
14
|
+
### End Users (Your Project)
|
|
14
15
|
|
|
15
|
-
|
|
16
|
+
End users access rules **only through MCP tools**. No local rule files needed.
|
|
17
|
+
|
|
18
|
+
```json
|
|
19
|
+
// .cursor/mcp.json
|
|
20
|
+
{
|
|
21
|
+
"mcpServers": {
|
|
22
|
+
"codingbuddy": {
|
|
23
|
+
"command": "npx",
|
|
24
|
+
"args": ["-y", "codingbuddy"]
|
|
25
|
+
}
|
|
26
|
+
}
|
|
27
|
+
}
|
|
28
|
+
```
|
|
29
|
+
|
|
30
|
+
Optional: Create `.cursor/rules/codingbuddy.mdc` for basic integration:
|
|
31
|
+
|
|
32
|
+
```yaml
|
|
16
33
|
---
|
|
17
|
-
description:
|
|
34
|
+
description: codingbuddy integration
|
|
18
35
|
globs:
|
|
19
36
|
alwaysApply: true
|
|
20
37
|
---
|
|
21
38
|
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
This project uses shared rules from `.ai-rules/` directory for all AI assistants.
|
|
25
|
-
|
|
26
|
-
## 📚 Core Rules
|
|
27
|
-
See [../../.ai-rules/rules/core.md](../../.ai-rules/rules/core.md) for:
|
|
28
|
-
- PLAN/ACT/EVAL workflow modes
|
|
29
|
-
- Agent activation rules
|
|
30
|
-
- Communication guidelines
|
|
31
|
-
|
|
32
|
-
## 🏗️ Project Setup
|
|
33
|
-
See [../../.ai-rules/rules/project.md](../../.ai-rules/rules/project.md) for:
|
|
34
|
-
- Tech stack and dependencies
|
|
35
|
-
- Project structure and architecture
|
|
36
|
-
- Development rules and conventions
|
|
37
|
-
- Domain knowledge and business context
|
|
38
|
-
|
|
39
|
-
## 🎯 Augmented Coding Principles
|
|
40
|
-
See [../../.ai-rules/rules/augmented-coding.md](../../.ai-rules/rules/augmented-coding.md) for:
|
|
41
|
-
- TDD cycle (Red → Green → Refactor)
|
|
42
|
-
- Code quality standards (SOLID, DRY)
|
|
43
|
-
- Testing best practices
|
|
44
|
-
- Commit discipline
|
|
45
|
-
|
|
46
|
-
## 🤖 Specialist Agents
|
|
47
|
-
See [../../.ai-rules/agents/README.md](../../.ai-rules/agents/README.md) for available specialist agents:
|
|
48
|
-
- Frontend Developer, Code Reviewer
|
|
49
|
-
- Architecture, Test Strategy, Performance, Security
|
|
50
|
-
- Accessibility, SEO, Design System, Documentation
|
|
51
|
-
- Code Quality, DevOps Engineer
|
|
39
|
+
When PLAN, ACT, EVAL keywords detected → call `parse_mode` MCP tool
|
|
52
40
|
```
|
|
53
41
|
|
|
54
|
-
###
|
|
42
|
+
### Monorepo Contributors
|
|
55
43
|
|
|
56
|
-
|
|
44
|
+
Contributors to the codingbuddy repository can use direct file references:
|
|
57
45
|
|
|
58
|
-
```
|
|
46
|
+
```
|
|
47
|
+
Project Root/
|
|
48
|
+
├── AGENTS.md # Cross-platform entry point
|
|
49
|
+
├── .cursor/rules/
|
|
50
|
+
│ ├── imports.mdc # Common rules (alwaysApply: true)
|
|
51
|
+
│ ├── auto-agent.mdc # File pattern-based Agent auto-activation
|
|
52
|
+
│ └── custom.mdc # Personal settings (Git ignored)
|
|
53
|
+
└── packages/rules/.ai-rules/ # Single Source of Truth
|
|
54
|
+
```
|
|
55
|
+
|
|
56
|
+
## DRY Principle
|
|
57
|
+
|
|
58
|
+
**Single Source of Truth**: `packages/rules/.ai-rules/`
|
|
59
|
+
|
|
60
|
+
- All Agent definitions, rules, skills managed only in `.ai-rules/`
|
|
61
|
+
- AGENTS.md and .mdc files act as **pointers only**
|
|
62
|
+
- No duplication, only references
|
|
63
|
+
|
|
64
|
+
## Configuration Files
|
|
65
|
+
|
|
66
|
+
### imports.mdc (alwaysApply)
|
|
67
|
+
|
|
68
|
+
Core rules automatically applied to all conversations:
|
|
69
|
+
|
|
70
|
+
```yaml
|
|
59
71
|
---
|
|
60
|
-
description:
|
|
72
|
+
description: codingbuddy common rules
|
|
61
73
|
globs:
|
|
62
74
|
alwaysApply: true
|
|
63
75
|
---
|
|
64
76
|
|
|
65
|
-
#
|
|
77
|
+
# Core principles only (details in .ai-rules/)
|
|
78
|
+
```
|
|
66
79
|
|
|
67
|
-
|
|
80
|
+
### auto-agent.mdc (glob-based)
|
|
68
81
|
|
|
69
|
-
|
|
82
|
+
Automatically provides appropriate Agent context based on file patterns:
|
|
70
83
|
|
|
71
|
-
|
|
84
|
+
```yaml
|
|
85
|
+
---
|
|
86
|
+
description: Agent auto-activation
|
|
87
|
+
globs:
|
|
88
|
+
- "**/*.tsx"
|
|
89
|
+
- "**/*.ts"
|
|
90
|
+
- "**/*.go"
|
|
91
|
+
alwaysApply: false
|
|
92
|
+
---
|
|
72
93
|
|
|
73
|
-
|
|
94
|
+
# File pattern → Agent mapping table
|
|
74
95
|
```
|
|
75
96
|
|
|
76
|
-
##
|
|
97
|
+
## Usage
|
|
98
|
+
|
|
99
|
+
### Mode Keywords
|
|
77
100
|
|
|
78
101
|
```
|
|
79
|
-
|
|
80
|
-
├── agents/ # Keep for Cursor compatibility
|
|
81
|
-
├── rules/
|
|
82
|
-
│ ├── core.mdc # Keep existing (can add reference to .ai-rules)
|
|
83
|
-
│ ├── project.mdc # Keep existing (can add reference to .ai-rules)
|
|
84
|
-
│ ├── augmented-coding.mdc # Keep existing
|
|
85
|
-
│ ├── imports.mdc # NEW: References to .ai-rules
|
|
86
|
-
│ └── cursor-specific.mdc # NEW: Cursor-only features
|
|
87
|
-
└── config.json # Cursor configuration
|
|
88
|
-
|
|
89
|
-
.ai-rules/ # Common rules for all AI tools
|
|
90
|
-
├── rules/
|
|
91
|
-
│ ├── core.md
|
|
92
|
-
│ ├── project.md
|
|
93
|
-
│ └── augmented-coding.md
|
|
94
|
-
├── agents/
|
|
95
|
-
│ └── *.json
|
|
96
|
-
└── adapters/
|
|
97
|
-
└── cursor.md (this file)
|
|
102
|
+
PLAN Design user authentication feature
|
|
98
103
|
```
|
|
99
104
|
|
|
100
|
-
|
|
105
|
+
→ `parse_mode` MCP tool is called, loading appropriate Agent and rules
|
|
101
106
|
|
|
102
|
-
###
|
|
107
|
+
### Auto-Activation on File Edit
|
|
103
108
|
|
|
104
|
-
|
|
105
|
-
```
|
|
106
|
-
@.ai-rules/rules/core.md
|
|
107
|
-
@.ai-rules/agents/frontend-developer.json
|
|
109
|
+
Open `.tsx` file → `auto-agent.mdc` auto-applies → frontend-developer Agent recommended
|
|
108
110
|
|
|
109
|
-
|
|
110
|
-
```
|
|
111
|
+
### Specialist Usage
|
|
111
112
|
|
|
112
|
-
|
|
113
|
-
|
|
114
|
-
|
|
113
|
+
```
|
|
114
|
+
EVAL Review from security perspective
|
|
115
|
+
```
|
|
115
116
|
|
|
116
|
-
|
|
117
|
+
→ security-specialist activated
|
|
117
118
|
|
|
118
|
-
|
|
119
|
-
- ✅ Access to common rules shared across all AI tools
|
|
120
|
-
- ✅ Cursor-specific features (globs, alwaysApply) still work
|
|
121
|
-
- ✅ Easy to update: change `.ai-rules/` once, all tools benefit
|
|
119
|
+
## MCP Tools
|
|
122
120
|
|
|
123
|
-
|
|
121
|
+
Available codingbuddy MCP tools in Cursor:
|
|
124
122
|
|
|
125
|
-
|
|
126
|
-
|
|
127
|
-
|
|
128
|
-
|
|
123
|
+
| Tool | Purpose |
|
|
124
|
+
|------|---------|
|
|
125
|
+
| `parse_mode` | Parse mode keywords + load Agent/rules |
|
|
126
|
+
| `get_agent_details` | Get specific Agent details |
|
|
127
|
+
| `get_project_config` | Get project configuration |
|
|
128
|
+
| `recommend_skills` | Recommend skills based on prompt |
|
|
129
|
+
| `prepare_parallel_agents` | Prepare parallel Agent execution |
|
|
129
130
|
|
|
130
131
|
## Skills
|
|
131
132
|
|
|
132
133
|
### Using Skills in Cursor
|
|
133
134
|
|
|
134
|
-
|
|
135
|
+
Load skills via file reference (monorepo only):
|
|
135
136
|
|
|
136
137
|
```
|
|
137
|
-
|
|
138
|
+
@packages/rules/.ai-rules/skills/test-driven-development/SKILL.md
|
|
138
139
|
```
|
|
139
140
|
|
|
140
|
-
|
|
141
|
+
For end users, use `recommend_skills` MCP tool instead.
|
|
141
142
|
|
|
142
143
|
### Available Skills
|
|
143
144
|
|
|
144
|
-
-
|
|
145
|
-
-
|
|
146
|
-
-
|
|
147
|
-
-
|
|
148
|
-
-
|
|
149
|
-
-
|
|
150
|
-
-
|
|
151
|
-
-
|
|
145
|
+
- `brainstorming/SKILL.md` - Idea → Design
|
|
146
|
+
- `test-driven-development/SKILL.md` - TDD workflow
|
|
147
|
+
- `systematic-debugging/SKILL.md` - Systematic debugging
|
|
148
|
+
- `writing-plans/SKILL.md` - Implementation plan writing
|
|
149
|
+
- `executing-plans/SKILL.md` - Plan execution
|
|
150
|
+
- `subagent-driven-development/SKILL.md` - Subagent development
|
|
151
|
+
- `dispatching-parallel-agents/SKILL.md` - Parallel Agent dispatch
|
|
152
|
+
- `frontend-design/SKILL.md` - Frontend design
|
|
153
|
+
|
|
154
|
+
## AGENTS.md
|
|
155
|
+
|
|
156
|
+
Industry standard format compatible with all AI tools (Cursor, Claude Code, Codex, etc.):
|
|
157
|
+
|
|
158
|
+
```markdown
|
|
159
|
+
# AGENTS.md
|
|
160
|
+
|
|
161
|
+
This project uses codingbuddy MCP server to manage AI Agents.
|
|
162
|
+
|
|
163
|
+
## Quick Start
|
|
164
|
+
...
|
|
165
|
+
```
|
|
166
|
+
|
|
167
|
+
See `AGENTS.md` in project root for details.
|
|
168
|
+
|
|
169
|
+
## Reference
|
|
170
|
+
|
|
171
|
+
- [AGENTS.md Official Spec](https://agents.md)
|
|
172
|
+
- [Cursor Rules Documentation](https://cursor.com/docs/context/rules)
|
|
173
|
+
- [codingbuddy MCP API](../../docs/api.md)
|
package/package.json
CHANGED