@neyugn/agent-kits 0.3.6 → 0.3.7
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.
|
@@ -49,9 +49,31 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
49
49
|
|
|
50
50
|
1. **Analyze**: Detect domains (Frontend, Backend, Security, etc.)
|
|
51
51
|
2. **Select**: Choose appropriate specialist(s)
|
|
52
|
-
3.
|
|
52
|
+
3. **🔴 Announce**: Your **VERY FIRST line** of response MUST be: `⚡ **@[agent-name] activated!**`
|
|
53
53
|
4. **Apply**: Use agent's persona and rules
|
|
54
54
|
|
|
55
|
+
> 🔴 **MANDATORY ANNOUNCEMENT RULE:**
|
|
56
|
+
> - You MUST output the announcement as the **first line** of every response before ANY other text.
|
|
57
|
+
> - Format: `⚡ **@agent-name activated!**` (replace `agent-name` with actual agent slug)
|
|
58
|
+
> - For multi-agent tasks: announce each agent on separate lines
|
|
59
|
+
> - This is NON-NEGOTIABLE. The user RELIES on this to verify correct agent routing.
|
|
60
|
+
>
|
|
61
|
+
> **Example — Single agent:**
|
|
62
|
+
> ```
|
|
63
|
+
> ⚡ **@backend-specialist activated!**
|
|
64
|
+
>
|
|
65
|
+
> Let me analyze your API endpoint...
|
|
66
|
+
> ```
|
|
67
|
+
>
|
|
68
|
+
> **Example — Multiple agents:**
|
|
69
|
+
> ```
|
|
70
|
+
> ⚡ **@orchestrator activated!**
|
|
71
|
+
> ⚡ **@frontend-specialist activated!**
|
|
72
|
+
> ⚡ **@backend-specialist activated!**
|
|
73
|
+
>
|
|
74
|
+
> I'll coordinate the full-stack implementation...
|
|
75
|
+
> ```
|
|
76
|
+
|
|
55
77
|
### Tier 1: Master Agents
|
|
56
78
|
|
|
57
79
|
| Agent | Use When |
|
|
@@ -105,11 +127,12 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
105
127
|
| ---- | ------------------------------- | --------------------------------- |
|
|
106
128
|
| 1 | Correct agent identified? | → Analyze domain |
|
|
107
129
|
| 2 | Read agent's .md file? | → Open `.agent/agents/{agent}.md` |
|
|
108
|
-
| 3 | Announced @agent?
|
|
130
|
+
| 3 | Announced @agent as FIRST LINE? | → 🔴 Add announcement IMMEDIATELY |
|
|
109
131
|
| 4 | Loaded skills from frontmatter? | → Check `skills:` field |
|
|
110
132
|
|
|
111
133
|
❌ Code without agent = PROTOCOL VIOLATION
|
|
112
134
|
❌ Skip announcement = USER CANNOT VERIFY
|
|
135
|
+
❌ Announcement NOT as first line = PROTOCOL VIOLATION
|
|
113
136
|
|
|
114
137
|
---
|
|
115
138
|
|
|
@@ -49,9 +49,31 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
49
49
|
|
|
50
50
|
1. **Analyze**: Detect domains (Frontend, Backend, Security, etc.)
|
|
51
51
|
2. **Select**: Choose appropriate specialist(s)
|
|
52
|
-
3.
|
|
52
|
+
3. **🔴 Announce**: Your **VERY FIRST line** of response MUST be: `⚡ **@[agent-name] activated!**`
|
|
53
53
|
4. **Apply**: Use agent's persona and rules
|
|
54
54
|
|
|
55
|
+
> 🔴 **MANDATORY ANNOUNCEMENT RULE:**
|
|
56
|
+
> - You MUST output the announcement as the **first line** of every response before ANY other text.
|
|
57
|
+
> - Format: `⚡ **@agent-name activated!**` (replace `agent-name` with actual agent slug)
|
|
58
|
+
> - For multi-agent tasks: announce each agent on separate lines
|
|
59
|
+
> - This is NON-NEGOTIABLE. The user RELIES on this to verify correct agent routing.
|
|
60
|
+
>
|
|
61
|
+
> **Example — Single agent:**
|
|
62
|
+
> ```
|
|
63
|
+
> ⚡ **@backend-specialist activated!**
|
|
64
|
+
>
|
|
65
|
+
> Let me analyze your API endpoint...
|
|
66
|
+
> ```
|
|
67
|
+
>
|
|
68
|
+
> **Example — Multiple agents:**
|
|
69
|
+
> ```
|
|
70
|
+
> ⚡ **@orchestrator activated!**
|
|
71
|
+
> ⚡ **@frontend-specialist activated!**
|
|
72
|
+
> ⚡ **@backend-specialist activated!**
|
|
73
|
+
>
|
|
74
|
+
> I'll coordinate the full-stack implementation...
|
|
75
|
+
> ```
|
|
76
|
+
|
|
55
77
|
### Tier 1: Master Agents
|
|
56
78
|
|
|
57
79
|
| Agent | Use When |
|
|
@@ -105,11 +127,12 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
105
127
|
| ---- | ------------------------------- | --------------------------------- |
|
|
106
128
|
| 1 | Correct agent identified? | → Analyze domain |
|
|
107
129
|
| 2 | Read agent's .md file? | → Open `.agent/agents/{agent}.md` |
|
|
108
|
-
| 3 | Announced @agent?
|
|
130
|
+
| 3 | Announced @agent as FIRST LINE? | → 🔴 Add announcement IMMEDIATELY |
|
|
109
131
|
| 4 | Loaded skills from frontmatter? | → Check `skills:` field |
|
|
110
132
|
|
|
111
133
|
❌ Code without agent = PROTOCOL VIOLATION
|
|
112
134
|
❌ Skip announcement = USER CANNOT VERIFY
|
|
135
|
+
❌ Announcement NOT as first line = PROTOCOL VIOLATION
|
|
113
136
|
|
|
114
137
|
---
|
|
115
138
|
|
|
@@ -55,9 +55,31 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
55
55
|
|
|
56
56
|
1. **Analyze**: Detect domains (Frontend, Backend, Security, etc.)
|
|
57
57
|
2. **Select**: Choose appropriate specialist(s)
|
|
58
|
-
3.
|
|
58
|
+
3. **🔴 Announce**: Your **VERY FIRST line** of response MUST be: `⚡ **@[agent-name] activated!**`
|
|
59
59
|
4. **Apply**: Use agent's persona and rules
|
|
60
60
|
|
|
61
|
+
> 🔴 **MANDATORY ANNOUNCEMENT RULE:**
|
|
62
|
+
> - You MUST output the announcement as the **first line** of every response before ANY other text.
|
|
63
|
+
> - Format: `⚡ **@agent-name activated!**` (replace `agent-name` with actual agent slug)
|
|
64
|
+
> - For multi-agent tasks: announce each agent on separate lines
|
|
65
|
+
> - This is NON-NEGOTIABLE. The user RELIES on this to verify correct agent routing.
|
|
66
|
+
>
|
|
67
|
+
> **Example — Single agent:**
|
|
68
|
+
> ```
|
|
69
|
+
> ⚡ **@backend-specialist activated!**
|
|
70
|
+
>
|
|
71
|
+
> Let me analyze your API endpoint...
|
|
72
|
+
> ```
|
|
73
|
+
>
|
|
74
|
+
> **Example — Multiple agents:**
|
|
75
|
+
> ```
|
|
76
|
+
> ⚡ **@orchestrator activated!**
|
|
77
|
+
> ⚡ **@frontend-specialist activated!**
|
|
78
|
+
> ⚡ **@backend-specialist activated!**
|
|
79
|
+
>
|
|
80
|
+
> I'll coordinate the full-stack implementation...
|
|
81
|
+
> ```
|
|
82
|
+
|
|
61
83
|
### Tier 1: Master Agents
|
|
62
84
|
|
|
63
85
|
| Agent | Use When |
|
|
@@ -111,11 +133,12 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
111
133
|
| ---- | ------------------------------- | --------------------------------- |
|
|
112
134
|
| 1 | Correct agent identified? | → Analyze domain |
|
|
113
135
|
| 2 | Read agent's .md file? | → Open `.agent/agents/{agent}.md` |
|
|
114
|
-
| 3 | Announced @agent?
|
|
136
|
+
| 3 | Announced @agent as FIRST LINE? | → 🔴 Add announcement IMMEDIATELY |
|
|
115
137
|
| 4 | Loaded skills from frontmatter? | → Check `skills:` field |
|
|
116
138
|
|
|
117
139
|
❌ Code without agent = PROTOCOL VIOLATION
|
|
118
140
|
❌ Skip announcement = USER CANNOT VERIFY
|
|
141
|
+
❌ Announcement NOT as first line = PROTOCOL VIOLATION
|
|
119
142
|
|
|
120
143
|
---
|
|
121
144
|
|
|
@@ -53,9 +53,31 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
53
53
|
|
|
54
54
|
1. **Analyze**: Detect domains (Frontend, Backend, Security, etc.)
|
|
55
55
|
2. **Select**: Choose appropriate specialist(s)
|
|
56
|
-
3.
|
|
56
|
+
3. **🔴 Announce**: Your **VERY FIRST line** of response MUST be: `⚡ **@[agent-name] activated!**`
|
|
57
57
|
4. **Apply**: Use agent's persona and rules
|
|
58
58
|
|
|
59
|
+
> 🔴 **MANDATORY ANNOUNCEMENT RULE:**
|
|
60
|
+
> - You MUST output the announcement as the **first line** of every response before ANY other text.
|
|
61
|
+
> - Format: `⚡ **@agent-name activated!**` (replace `agent-name` with actual agent slug)
|
|
62
|
+
> - For multi-agent tasks: announce each agent on separate lines
|
|
63
|
+
> - This is NON-NEGOTIABLE. The user RELIES on this to verify correct agent routing.
|
|
64
|
+
>
|
|
65
|
+
> **Example — Single agent:**
|
|
66
|
+
> ```
|
|
67
|
+
> ⚡ **@backend-specialist activated!**
|
|
68
|
+
>
|
|
69
|
+
> Let me analyze your API endpoint...
|
|
70
|
+
> ```
|
|
71
|
+
>
|
|
72
|
+
> **Example — Multiple agents:**
|
|
73
|
+
> ```
|
|
74
|
+
> ⚡ **@orchestrator activated!**
|
|
75
|
+
> ⚡ **@frontend-specialist activated!**
|
|
76
|
+
> ⚡ **@backend-specialist activated!**
|
|
77
|
+
>
|
|
78
|
+
> I'll coordinate the full-stack implementation...
|
|
79
|
+
> ```
|
|
80
|
+
|
|
59
81
|
### Tier 1: Master Agents
|
|
60
82
|
|
|
61
83
|
| Agent | Use When |
|
|
@@ -109,11 +131,12 @@ Agent activated → Check frontmatter `skills:` → Read SKILL.md → Apply.
|
|
|
109
131
|
| ---- | ------------------------------- | --------------------------------- |
|
|
110
132
|
| 1 | Correct agent identified? | → Analyze domain |
|
|
111
133
|
| 2 | Read agent's .md file? | → Open `.agent/agents/{agent}.md` |
|
|
112
|
-
| 3 | Announced @agent?
|
|
134
|
+
| 3 | Announced @agent as FIRST LINE? | → 🔴 Add announcement IMMEDIATELY |
|
|
113
135
|
| 4 | Loaded skills from frontmatter? | → Check `skills:` field |
|
|
114
136
|
|
|
115
137
|
❌ Code without agent = PROTOCOL VIOLATION
|
|
116
138
|
❌ Skip announcement = USER CANNOT VERIFY
|
|
139
|
+
❌ Announcement NOT as first line = PROTOCOL VIOLATION
|
|
117
140
|
|
|
118
141
|
---
|
|
119
142
|
|
|
@@ -49,9 +49,31 @@ Agent activated → Check ARCHITECTURE.md for assigned skills → Use `skill` to
|
|
|
49
49
|
|
|
50
50
|
1. **Analyze**: Detect domains (Frontend, Backend, Security, etc.)
|
|
51
51
|
2. **Select**: Choose appropriate specialist(s)
|
|
52
|
-
3.
|
|
52
|
+
3. **🔴 Announce**: Your **VERY FIRST line** of response MUST be: `⚡ **@[agent-name] activated!**`
|
|
53
53
|
4. **Apply**: Use agent's persona and rules
|
|
54
54
|
|
|
55
|
+
> 🔴 **MANDATORY ANNOUNCEMENT RULE:**
|
|
56
|
+
> - You MUST output the announcement as the **first line** of every response before ANY other text.
|
|
57
|
+
> - Format: `⚡ **@agent-name activated!**` (replace `agent-name` with actual agent slug)
|
|
58
|
+
> - For multi-agent tasks: announce each agent on separate lines
|
|
59
|
+
> - This is NON-NEGOTIABLE. The user RELIES on this to verify correct agent routing.
|
|
60
|
+
>
|
|
61
|
+
> **Example — Single agent:**
|
|
62
|
+
> ```
|
|
63
|
+
> ⚡ **@backend-specialist activated!**
|
|
64
|
+
>
|
|
65
|
+
> Let me analyze your API endpoint...
|
|
66
|
+
> ```
|
|
67
|
+
>
|
|
68
|
+
> **Example — Multiple agents:**
|
|
69
|
+
> ```
|
|
70
|
+
> ⚡ **@orchestrator activated!**
|
|
71
|
+
> ⚡ **@frontend-specialist activated!**
|
|
72
|
+
> ⚡ **@backend-specialist activated!**
|
|
73
|
+
>
|
|
74
|
+
> I'll coordinate the full-stack implementation...
|
|
75
|
+
> ```
|
|
76
|
+
|
|
55
77
|
### Tier 1: Master Agents
|
|
56
78
|
|
|
57
79
|
| Agent | Use When |
|
|
@@ -105,11 +127,12 @@ Agent activated → Check ARCHITECTURE.md for assigned skills → Use `skill` to
|
|
|
105
127
|
| ---- | --------------------------------- | ------------------------------------------- |
|
|
106
128
|
| 1 | Correct agent identified? | → Analyze domain |
|
|
107
129
|
| 2 | Read agent's .md file? | → Open `.agent/agents/{agent}.md` |
|
|
108
|
-
| 3 | Announced @agent?
|
|
130
|
+
| 3 | Announced @agent as FIRST LINE? | → 🔴 Add announcement IMMEDIATELY |
|
|
109
131
|
| 4 | Loaded skills from ARCHITECTURE? | → Check ARCHITECTURE.md agent-skills table |
|
|
110
132
|
|
|
111
133
|
❌ Code without agent = PROTOCOL VIOLATION
|
|
112
134
|
❌ Skip announcement = USER CANNOT VERIFY
|
|
135
|
+
❌ Announcement NOT as first line = PROTOCOL VIOLATION
|
|
113
136
|
|
|
114
137
|
---
|
|
115
138
|
|