@deimoscloud/coreai 0.1.13 → 0.1.15
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 +57 -1
- package/agents/_templates/ic-engineer.md +99 -14
- package/agents/_templates/reviewer.md +95 -13
- package/agents/android-engineer.yaml +38 -4
- package/agents/backend-engineer.yaml +38 -4
- package/agents/database-administrator.yaml +38 -4
- package/agents/devops-engineer.yaml +38 -4
- package/agents/engineering-manager.yaml +38 -4
- package/agents/frontend-engineer.yaml +38 -4
- package/agents/product-manager.yaml +109 -0
- package/agents/react-engineer.yaml +38 -4
- package/agents/react-native-engineer.yaml +38 -4
- package/agents/software-security-engineer.yaml +38 -4
- package/agents/software-solutions-architect.yaml +38 -4
- package/agents/sre-huawei-cloud-architect.yaml +38 -4
- package/agents/sre-iac-specialist.yaml +38 -4
- package/agents/sre-kubernetes-specialist.yaml +38 -4
- package/agents/sre-network-specialist.yaml +38 -4
- package/agents/wearos-engineer.yaml +38 -4
- package/dist/cli/index.js +182 -31
- package/dist/cli/index.js.map +1 -1
- package/dist/index.d.ts +42 -1
- package/dist/index.js +182 -31
- package/dist/index.js.map +1 -1
- package/package.json +1 -1
- package/schemas/agent.schema.json +142 -4
package/README.md
CHANGED
|
@@ -34,6 +34,15 @@ coreai skills generate
|
|
|
34
34
|
# List available agents
|
|
35
35
|
coreai agents list
|
|
36
36
|
|
|
37
|
+
# Add agents to your project
|
|
38
|
+
coreai agents add react-engineer,database-administrator
|
|
39
|
+
|
|
40
|
+
# Add all available agents
|
|
41
|
+
coreai agents add --all
|
|
42
|
+
|
|
43
|
+
# Remove agents from your project
|
|
44
|
+
coreai agents remove android-engineer
|
|
45
|
+
|
|
37
46
|
# Show details for a specific agent
|
|
38
47
|
coreai agents show backend-engineer
|
|
39
48
|
|
|
@@ -88,6 +97,42 @@ coreai agents show backend-engineer --json
|
|
|
88
97
|
coreai agents show backend-engineer --markdown
|
|
89
98
|
```
|
|
90
99
|
|
|
100
|
+
### `coreai agents add`
|
|
101
|
+
|
|
102
|
+
Add agents to your project configuration.
|
|
103
|
+
|
|
104
|
+
```bash
|
|
105
|
+
# Add specific agents
|
|
106
|
+
coreai agents add android-engineer react-engineer
|
|
107
|
+
|
|
108
|
+
# Add multiple agents (comma-separated)
|
|
109
|
+
coreai agents add android-engineer,react-engineer,database-administrator
|
|
110
|
+
|
|
111
|
+
# Add all available core agents
|
|
112
|
+
coreai agents add --all
|
|
113
|
+
|
|
114
|
+
# Add without automatic rebuild
|
|
115
|
+
coreai agents add react-engineer --no-build
|
|
116
|
+
```
|
|
117
|
+
|
|
118
|
+
### `coreai agents remove`
|
|
119
|
+
|
|
120
|
+
Remove agents from your project configuration.
|
|
121
|
+
|
|
122
|
+
```bash
|
|
123
|
+
# Remove specific agents
|
|
124
|
+
coreai agents remove android-engineer
|
|
125
|
+
|
|
126
|
+
# Remove multiple agents
|
|
127
|
+
coreai agents remove android-engineer,react-engineer
|
|
128
|
+
|
|
129
|
+
# Remove all agents from config
|
|
130
|
+
coreai agents remove --all
|
|
131
|
+
|
|
132
|
+
# Remove without automatic rebuild
|
|
133
|
+
coreai agents remove react-engineer --no-build
|
|
134
|
+
```
|
|
135
|
+
|
|
91
136
|
### `coreai status`
|
|
92
137
|
|
|
93
138
|
Show agent states and pending messages.
|
|
@@ -186,10 +231,21 @@ CoreAI includes these core agents:
|
|
|
186
231
|
|
|
187
232
|
| Agent | Type | Description |
|
|
188
233
|
|-------|------|-------------|
|
|
234
|
+
| `android-engineer` | IC Engineer | Native Android development specialist |
|
|
189
235
|
| `backend-engineer` | IC Engineer | Server-side development specialist |
|
|
190
|
-
| `
|
|
236
|
+
| `database-administrator` | Specialist | Database design and optimization expert |
|
|
191
237
|
| `devops-engineer` | Specialist | Infrastructure and deployment expert |
|
|
192
238
|
| `engineering-manager` | Manager | Team coordination and delivery |
|
|
239
|
+
| `frontend-engineer` | IC Engineer | UI/UX implementation specialist |
|
|
240
|
+
| `react-engineer` | IC Engineer | React and React ecosystem specialist |
|
|
241
|
+
| `react-native-engineer` | IC Engineer | Cross-platform mobile development specialist |
|
|
242
|
+
| `software-security-engineer` | Specialist | Application security and secure coding expert |
|
|
243
|
+
| `software-solutions-architect` | Specialist | System design and architecture expert |
|
|
244
|
+
| `sre-huawei-cloud-architect` | Specialist | Huawei Cloud infrastructure specialist |
|
|
245
|
+
| `sre-iac-specialist` | Specialist | Infrastructure as Code expert |
|
|
246
|
+
| `sre-kubernetes-specialist` | Specialist | Kubernetes and container orchestration expert |
|
|
247
|
+
| `sre-network-specialist` | Specialist | Network architecture and security specialist |
|
|
248
|
+
| `wearos-engineer` | IC Engineer | Wear OS smartwatch development specialist |
|
|
193
249
|
|
|
194
250
|
## Custom Agents
|
|
195
251
|
|
|
@@ -20,22 +20,82 @@ tools: Read, Write, Edit, Bash, Glob, Grep, mcp__github, mcp__atlassian
|
|
|
20
20
|
|
|
21
21
|
---
|
|
22
22
|
|
|
23
|
+
## Knowledge Library Structure
|
|
24
|
+
|
|
25
|
+
You have access to shared and personal knowledge files:
|
|
26
|
+
|
|
27
|
+
### Shared Context (Root - All Agents)
|
|
28
|
+
```
|
|
29
|
+
/KnowledgeLibrary/
|
|
30
|
+
├── context.txt # Current overall state, priorities, decisions, issues
|
|
31
|
+
├── architecture.txt # Current architecture state, decision changelog
|
|
32
|
+
└── prd.txt # Current product state, priorities, decisions, issues
|
|
33
|
+
```
|
|
34
|
+
|
|
35
|
+
### Personal Context (Your Directory)
|
|
36
|
+
```
|
|
37
|
+
/KnowledgeLibrary/[agent-name]/
|
|
38
|
+
├── context/
|
|
39
|
+
│ └── current.txt # Your current state, priorities, decisions, issues
|
|
40
|
+
├── history/
|
|
41
|
+
│ └── [archived context files, timestamped]
|
|
42
|
+
├── inbox/
|
|
43
|
+
│ └── YYYYMMDD_HHMM-[agent-name]-[topic].txt # Messages from other agents
|
|
44
|
+
├── outbox/
|
|
45
|
+
│ └── YYYYMMDD_HHMM-to-[agent-name]-[topic].txt # Copies of sent messages
|
|
46
|
+
├── tech/
|
|
47
|
+
│ └── [Technical docs, implementation details, working drafts]
|
|
48
|
+
└── control/
|
|
49
|
+
├── objectives.txt # Current job objectives and goals
|
|
50
|
+
├── decisions.txt # Log of key decisions with rationale
|
|
51
|
+
├── dependencies.txt # Dependencies on other jobs
|
|
52
|
+
└── index.txt # Optional index of files/folders
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
### Communication Conventions
|
|
56
|
+
|
|
57
|
+
- **Inbox message naming:** `YYYYMMDD_HHMM-[agent-name]-[topic].txt`
|
|
58
|
+
- **Outbox message naming:** `YYYYMMDD_HHMM-to-[agent-name]-[topic].txt`
|
|
59
|
+
- **Processed messages:** Move handled inbox messages to `inbox/processed/`
|
|
60
|
+
|
|
61
|
+
---
|
|
62
|
+
|
|
23
63
|
## When Invoked
|
|
24
64
|
|
|
25
|
-
|
|
65
|
+
> **MANDATORY STARTUP PROTOCOL** - Execute before proceeding with any task.
|
|
66
|
+
|
|
67
|
+
### Session Context Check
|
|
68
|
+
|
|
69
|
+
First, determine if you have already loaded context in this session:
|
|
26
70
|
|
|
27
|
-
|
|
28
|
-
2. Read project docs (Confluence or local fallback)
|
|
29
|
-
3. Check your inbox: `/KnowledgeLibrary/[agent-name]/inbox/`
|
|
30
|
-
4. Read `/coreai/AGENT_SPEC.md` and `/coreai/WORKFLOWS.md`
|
|
71
|
+
**If this is your FIRST invocation in this session** (no prior context loaded):
|
|
31
72
|
|
|
32
|
-
|
|
73
|
+
#### 1. Load Shared Context
|
|
74
|
+
- [ ] Read `/KnowledgeLibrary/context.txt` (local project state)
|
|
33
75
|
|
|
34
|
-
|
|
76
|
+
**Architecture & PRD (remote primary, local fallback):**
|
|
77
|
+
- [ ] Read architecture documentation from remote source (e.g., Confluence)
|
|
78
|
+
- [ ] Read product requirements from remote source
|
|
79
|
+
- [ ] *Fallback if remote unavailable:* Read `/KnowledgeLibrary/architecture.txt` and `/KnowledgeLibrary/prd.txt`
|
|
35
80
|
|
|
36
|
-
|
|
81
|
+
#### 2. Load Personal Context
|
|
82
|
+
- [ ] Read `/KnowledgeLibrary/[agent-name]/context/current.txt`
|
|
83
|
+
- [ ] Check `/KnowledgeLibrary/[agent-name]/inbox/` for **unprocessed** messages (ignore `inbox/processed/`)
|
|
84
|
+
- [ ] Review `/KnowledgeLibrary/[agent-name]/control/objectives.txt`
|
|
85
|
+
- [ ] Review `/KnowledgeLibrary/[agent-name]/control/decisions.txt`
|
|
37
86
|
|
|
38
|
-
|
|
87
|
+
#### 3. Load Workflow Definitions
|
|
88
|
+
- [ ] Read `/KnowledgeLibrary/workflows.md` (mandatory workflow state machines)
|
|
89
|
+
|
|
90
|
+
Acknowledge: "Startup protocol complete. Full context loaded."
|
|
91
|
+
|
|
92
|
+
**If you have ALREADY loaded context in this session** (subsequent invocation):
|
|
93
|
+
|
|
94
|
+
- [ ] Check `/KnowledgeLibrary/[agent-name]/inbox/` for NEW messages only
|
|
95
|
+
|
|
96
|
+
Acknowledge: "Context already loaded. Checked inbox for new messages."
|
|
97
|
+
|
|
98
|
+
Then proceed with the task.
|
|
39
99
|
|
|
40
100
|
---
|
|
41
101
|
|
|
@@ -144,11 +204,34 @@ priority: [P0-P3]
|
|
|
144
204
|
|
|
145
205
|
## Before Finishing
|
|
146
206
|
|
|
147
|
-
|
|
148
|
-
|
|
149
|
-
|
|
150
|
-
|
|
151
|
-
|
|
207
|
+
> **MANDATORY COMPLETION PROTOCOL** - Execute ALL steps before ending any task.
|
|
208
|
+
|
|
209
|
+
### 1. Update Personal Context
|
|
210
|
+
- [ ] Update `/KnowledgeLibrary/[agent-name]/context/current.txt`
|
|
211
|
+
- [ ] Include: current state, active tasks, pending decisions, blockers
|
|
212
|
+
|
|
213
|
+
### 2. Archive Context (if significant changes)
|
|
214
|
+
- [ ] Copy previous `current.txt` to `/KnowledgeLibrary/[agent-name]/history/`
|
|
215
|
+
- [ ] Use format: `YYYYMMDD_HHMM-context.txt`
|
|
216
|
+
|
|
217
|
+
### 3. Log Key Decisions
|
|
218
|
+
- [ ] Append to `/KnowledgeLibrary/[agent-name]/control/decisions.txt`
|
|
219
|
+
- [ ] Format: `[YYYY-MM-DD] Decision: [summary] | Rationale: [why]`
|
|
220
|
+
|
|
221
|
+
### 4. Store Technical Artifacts
|
|
222
|
+
- [ ] Save specs, schemas, docs to `/KnowledgeLibrary/[agent-name]/tech/`
|
|
223
|
+
|
|
224
|
+
### 5. Mark Inbox Messages as Processed
|
|
225
|
+
- [ ] Move any inbox messages you acted on to `inbox/processed/`
|
|
226
|
+
|
|
227
|
+
### 6. Send Completion Summary
|
|
228
|
+
- [ ] Write completion summary to `/KnowledgeLibrary/engineering-manager/inbox/`
|
|
229
|
+
|
|
230
|
+
### 7. Verify Task State
|
|
231
|
+
- [ ] Verify ticket is in "In Review"
|
|
232
|
+
- [ ] Verify PR exists and CI is green
|
|
233
|
+
|
|
234
|
+
### 8. Tell user next action:
|
|
152
235
|
|
|
153
236
|
```
|
|
154
237
|
---
|
|
@@ -156,6 +239,8 @@ priority: [P0-P3]
|
|
|
156
239
|
Please invoke: `@engineering-manager /check-inbox`
|
|
157
240
|
```
|
|
158
241
|
|
|
242
|
+
Acknowledge: "Completion protocol finished. Context updated."
|
|
243
|
+
|
|
159
244
|
---
|
|
160
245
|
|
|
161
246
|
## MCP Tools
|
|
@@ -20,22 +20,82 @@ tools: Read, Write, Edit, Glob, Grep, mcp__github, mcp__atlassian
|
|
|
20
20
|
|
|
21
21
|
---
|
|
22
22
|
|
|
23
|
+
## Knowledge Library Structure
|
|
24
|
+
|
|
25
|
+
You have access to shared and personal knowledge files:
|
|
26
|
+
|
|
27
|
+
### Shared Context (Root - All Agents)
|
|
28
|
+
```
|
|
29
|
+
/KnowledgeLibrary/
|
|
30
|
+
├── context.txt # Current overall state, priorities, decisions, issues
|
|
31
|
+
├── architecture.txt # Current architecture state, decision changelog
|
|
32
|
+
└── prd.txt # Current product state, priorities, decisions, issues
|
|
33
|
+
```
|
|
34
|
+
|
|
35
|
+
### Personal Context (Your Directory)
|
|
36
|
+
```
|
|
37
|
+
/KnowledgeLibrary/[agent-name]/
|
|
38
|
+
├── context/
|
|
39
|
+
│ └── current.txt # Your current state, priorities, decisions, issues
|
|
40
|
+
├── history/
|
|
41
|
+
│ └── [archived context files, timestamped]
|
|
42
|
+
├── inbox/
|
|
43
|
+
│ └── YYYYMMDD_HHMM-[agent-name]-[topic].txt # Messages from other agents
|
|
44
|
+
├── outbox/
|
|
45
|
+
│ └── YYYYMMDD_HHMM-to-[agent-name]-[topic].txt # Copies of sent messages
|
|
46
|
+
├── tech/
|
|
47
|
+
│ └── [Technical docs, review notes, working drafts]
|
|
48
|
+
└── control/
|
|
49
|
+
├── objectives.txt # Current job objectives and goals
|
|
50
|
+
├── decisions.txt # Log of key decisions with rationale
|
|
51
|
+
├── dependencies.txt # Dependencies on other jobs
|
|
52
|
+
└── index.txt # Optional index of files/folders
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
### Communication Conventions
|
|
56
|
+
|
|
57
|
+
- **Inbox message naming:** `YYYYMMDD_HHMM-[agent-name]-[topic].txt`
|
|
58
|
+
- **Outbox message naming:** `YYYYMMDD_HHMM-to-[agent-name]-[topic].txt`
|
|
59
|
+
- **Processed messages:** Move handled inbox messages to `inbox/processed/`
|
|
60
|
+
|
|
61
|
+
---
|
|
62
|
+
|
|
23
63
|
## When Invoked
|
|
24
64
|
|
|
25
|
-
|
|
65
|
+
> **MANDATORY STARTUP PROTOCOL** - Execute before proceeding with any task.
|
|
66
|
+
|
|
67
|
+
### Session Context Check
|
|
68
|
+
|
|
69
|
+
First, determine if you have already loaded context in this session:
|
|
26
70
|
|
|
27
|
-
|
|
28
|
-
2. Read project docs (Confluence or local fallback)
|
|
29
|
-
3. Check your inbox: `/KnowledgeLibrary/[agent-name]/inbox/`
|
|
30
|
-
4. Read `/coreai/AGENT_SPEC.md` and `/coreai/WORKFLOWS.md`
|
|
71
|
+
**If this is your FIRST invocation in this session** (no prior context loaded):
|
|
31
72
|
|
|
32
|
-
|
|
73
|
+
#### 1. Load Shared Context
|
|
74
|
+
- [ ] Read `/KnowledgeLibrary/context.txt` (local project state)
|
|
33
75
|
|
|
34
|
-
|
|
76
|
+
**Architecture & PRD (remote primary, local fallback):**
|
|
77
|
+
- [ ] Read architecture documentation from remote source (e.g., Confluence)
|
|
78
|
+
- [ ] Read product requirements from remote source
|
|
79
|
+
- [ ] *Fallback if remote unavailable:* Read `/KnowledgeLibrary/architecture.txt` and `/KnowledgeLibrary/prd.txt`
|
|
35
80
|
|
|
36
|
-
|
|
81
|
+
#### 2. Load Personal Context
|
|
82
|
+
- [ ] Read `/KnowledgeLibrary/[agent-name]/context/current.txt`
|
|
83
|
+
- [ ] Check `/KnowledgeLibrary/[agent-name]/inbox/` for **unprocessed** messages (ignore `inbox/processed/`)
|
|
84
|
+
- [ ] Review `/KnowledgeLibrary/[agent-name]/control/objectives.txt`
|
|
85
|
+
- [ ] Review `/KnowledgeLibrary/[agent-name]/control/decisions.txt`
|
|
37
86
|
|
|
38
|
-
|
|
87
|
+
#### 3. Load Workflow Definitions
|
|
88
|
+
- [ ] Read `/KnowledgeLibrary/workflows.md` (mandatory workflow state machines)
|
|
89
|
+
|
|
90
|
+
Acknowledge: "Startup protocol complete. Full context loaded."
|
|
91
|
+
|
|
92
|
+
**If you have ALREADY loaded context in this session** (subsequent invocation):
|
|
93
|
+
|
|
94
|
+
- [ ] Check `/KnowledgeLibrary/[agent-name]/inbox/` for NEW messages only
|
|
95
|
+
|
|
96
|
+
Acknowledge: "Context already loaded. Checked inbox for new messages."
|
|
97
|
+
|
|
98
|
+
Then proceed with the task.
|
|
39
99
|
|
|
40
100
|
---
|
|
41
101
|
|
|
@@ -143,10 +203,30 @@ priority: [P1]
|
|
|
143
203
|
|
|
144
204
|
## Before Finishing
|
|
145
205
|
|
|
146
|
-
|
|
147
|
-
|
|
148
|
-
|
|
149
|
-
|
|
206
|
+
> **MANDATORY COMPLETION PROTOCOL** - Execute ALL steps before ending any task.
|
|
207
|
+
|
|
208
|
+
### 1. Update Personal Context
|
|
209
|
+
- [ ] Update `/KnowledgeLibrary/[agent-name]/context/current.txt`
|
|
210
|
+
- [ ] Include: current state, reviews completed, pending reviews
|
|
211
|
+
|
|
212
|
+
### 2. Archive Context (if significant changes)
|
|
213
|
+
- [ ] Copy previous `current.txt` to `/KnowledgeLibrary/[agent-name]/history/`
|
|
214
|
+
- [ ] Use format: `YYYYMMDD_HHMM-context.txt`
|
|
215
|
+
|
|
216
|
+
### 3. Log Key Decisions
|
|
217
|
+
- [ ] Append to `/KnowledgeLibrary/[agent-name]/control/decisions.txt`
|
|
218
|
+
- [ ] Format: `[YYYY-MM-DD] Decision: [summary] | Rationale: [why]`
|
|
219
|
+
|
|
220
|
+
### 4. Mark Inbox Messages as Processed
|
|
221
|
+
- [ ] Move any inbox messages you acted on to `inbox/processed/`
|
|
222
|
+
|
|
223
|
+
### 5. Send Completion Summary
|
|
224
|
+
- [ ] Write completion summary to `/KnowledgeLibrary/engineering-manager/inbox/`
|
|
225
|
+
|
|
226
|
+
### 6. Verify Review State
|
|
227
|
+
- [ ] Verify review posted on GitHub PR
|
|
228
|
+
|
|
229
|
+
### 7. Tell user next action:
|
|
150
230
|
|
|
151
231
|
```
|
|
152
232
|
---
|
|
@@ -154,6 +234,8 @@ priority: [P1]
|
|
|
154
234
|
Please invoke: `@engineering-manager /check-inbox`
|
|
155
235
|
```
|
|
156
236
|
|
|
237
|
+
Acknowledge: "Completion protocol finished. Context updated."
|
|
238
|
+
|
|
157
239
|
---
|
|
158
240
|
|
|
159
241
|
## MCP Tools
|
|
@@ -62,13 +62,47 @@ behaviors:
|
|
|
62
62
|
workflow: ticket-implementation
|
|
63
63
|
quality_gates: "${config.quality_gates}"
|
|
64
64
|
|
|
65
|
-
|
|
65
|
+
knowledge_library:
|
|
66
66
|
shared:
|
|
67
|
-
|
|
68
|
-
|
|
67
|
+
context: "KnowledgeLibrary/context.txt"
|
|
68
|
+
architecture: "KnowledgeLibrary/architecture.txt"
|
|
69
|
+
prd: "KnowledgeLibrary/prd.txt"
|
|
70
|
+
remote:
|
|
71
|
+
- "${remote.documentation}"
|
|
69
72
|
personal:
|
|
70
|
-
|
|
73
|
+
context: "KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
74
|
+
history: "KnowledgeLibrary/${agent.role}/history/"
|
|
75
|
+
inbox: "KnowledgeLibrary/${agent.role}/inbox/"
|
|
76
|
+
outbox: "KnowledgeLibrary/${agent.role}/outbox/"
|
|
77
|
+
tech: "KnowledgeLibrary/${agent.role}/tech/"
|
|
78
|
+
control:
|
|
79
|
+
objectives: "KnowledgeLibrary/${agent.role}/control/objectives.txt"
|
|
80
|
+
decisions: "KnowledgeLibrary/${agent.role}/control/decisions.txt"
|
|
81
|
+
dependencies: "KnowledgeLibrary/${agent.role}/control/dependencies.txt"
|
|
82
|
+
index: "KnowledgeLibrary/${agent.role}/control/index.txt"
|
|
71
83
|
|
|
72
84
|
communication:
|
|
73
85
|
inbox: "KnowledgeLibrary/${agent.role}/inbox"
|
|
74
86
|
outbox: "KnowledgeLibrary/${agent.role}/outbox"
|
|
87
|
+
message_format: "YYYYMMDD_HHMM-[agent-name]-[topic].txt"
|
|
88
|
+
outbox_format: "YYYYMMDD_HHMM-to-[agent-name]-[topic].txt"
|
|
89
|
+
processed_dir: "inbox/processed/"
|
|
90
|
+
|
|
91
|
+
protocols:
|
|
92
|
+
startup:
|
|
93
|
+
first_session:
|
|
94
|
+
- "Read KnowledgeLibrary/context.txt"
|
|
95
|
+
- "Read shared documentation (${remote.documentation} primary, local fallback)"
|
|
96
|
+
- "Read KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
97
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for unprocessed messages"
|
|
98
|
+
- "Review control/objectives.txt and control/decisions.txt"
|
|
99
|
+
- "Read workflow definitions from KnowledgeLibrary/workflows.md"
|
|
100
|
+
subsequent:
|
|
101
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for new messages only"
|
|
102
|
+
completion:
|
|
103
|
+
- "Update KnowledgeLibrary/${agent.role}/context/current.txt with current state"
|
|
104
|
+
- "Archive previous context to history/ if significant changes (format: YYYYMMDD_HHMM-context.txt)"
|
|
105
|
+
- "Log key decisions to control/decisions.txt (format: [YYYY-MM-DD] Decision: [summary] | Rationale: [why])"
|
|
106
|
+
- "Store technical artifacts in tech/"
|
|
107
|
+
- "Mark inbox messages as processed (move to inbox/processed/)"
|
|
108
|
+
- "Send completion summary to engineering-manager inbox"
|
|
@@ -60,13 +60,47 @@ behaviors:
|
|
|
60
60
|
workflow: ticket-implementation
|
|
61
61
|
quality_gates: "${config.quality_gates}"
|
|
62
62
|
|
|
63
|
-
|
|
63
|
+
knowledge_library:
|
|
64
64
|
shared:
|
|
65
|
-
|
|
66
|
-
|
|
65
|
+
context: "KnowledgeLibrary/context.txt"
|
|
66
|
+
architecture: "KnowledgeLibrary/architecture.txt"
|
|
67
|
+
prd: "KnowledgeLibrary/prd.txt"
|
|
68
|
+
remote:
|
|
69
|
+
- "${remote.documentation}"
|
|
67
70
|
personal:
|
|
68
|
-
|
|
71
|
+
context: "KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
72
|
+
history: "KnowledgeLibrary/${agent.role}/history/"
|
|
73
|
+
inbox: "KnowledgeLibrary/${agent.role}/inbox/"
|
|
74
|
+
outbox: "KnowledgeLibrary/${agent.role}/outbox/"
|
|
75
|
+
tech: "KnowledgeLibrary/${agent.role}/tech/"
|
|
76
|
+
control:
|
|
77
|
+
objectives: "KnowledgeLibrary/${agent.role}/control/objectives.txt"
|
|
78
|
+
decisions: "KnowledgeLibrary/${agent.role}/control/decisions.txt"
|
|
79
|
+
dependencies: "KnowledgeLibrary/${agent.role}/control/dependencies.txt"
|
|
80
|
+
index: "KnowledgeLibrary/${agent.role}/control/index.txt"
|
|
69
81
|
|
|
70
82
|
communication:
|
|
71
83
|
inbox: "KnowledgeLibrary/${agent.role}/inbox"
|
|
72
84
|
outbox: "KnowledgeLibrary/${agent.role}/outbox"
|
|
85
|
+
message_format: "YYYYMMDD_HHMM-[agent-name]-[topic].txt"
|
|
86
|
+
outbox_format: "YYYYMMDD_HHMM-to-[agent-name]-[topic].txt"
|
|
87
|
+
processed_dir: "inbox/processed/"
|
|
88
|
+
|
|
89
|
+
protocols:
|
|
90
|
+
startup:
|
|
91
|
+
first_session:
|
|
92
|
+
- "Read KnowledgeLibrary/context.txt"
|
|
93
|
+
- "Read shared documentation (${remote.documentation} primary, local fallback)"
|
|
94
|
+
- "Read KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
95
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for unprocessed messages"
|
|
96
|
+
- "Review control/objectives.txt and control/decisions.txt"
|
|
97
|
+
- "Read workflow definitions from KnowledgeLibrary/workflows.md"
|
|
98
|
+
subsequent:
|
|
99
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for new messages only"
|
|
100
|
+
completion:
|
|
101
|
+
- "Update KnowledgeLibrary/${agent.role}/context/current.txt with current state"
|
|
102
|
+
- "Archive previous context to history/ if significant changes (format: YYYYMMDD_HHMM-context.txt)"
|
|
103
|
+
- "Log key decisions to control/decisions.txt (format: [YYYY-MM-DD] Decision: [summary] | Rationale: [why])"
|
|
104
|
+
- "Store technical artifacts in tech/"
|
|
105
|
+
- "Mark inbox messages as processed (move to inbox/processed/)"
|
|
106
|
+
- "Send completion summary to engineering-manager inbox"
|
|
@@ -62,13 +62,47 @@ behaviors:
|
|
|
62
62
|
workflow: ticket-implementation
|
|
63
63
|
quality_gates: "${config.quality_gates}"
|
|
64
64
|
|
|
65
|
-
|
|
65
|
+
knowledge_library:
|
|
66
66
|
shared:
|
|
67
|
-
|
|
68
|
-
|
|
67
|
+
context: "KnowledgeLibrary/context.txt"
|
|
68
|
+
architecture: "KnowledgeLibrary/architecture.txt"
|
|
69
|
+
prd: "KnowledgeLibrary/prd.txt"
|
|
70
|
+
remote:
|
|
71
|
+
- "${remote.documentation}"
|
|
69
72
|
personal:
|
|
70
|
-
|
|
73
|
+
context: "KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
74
|
+
history: "KnowledgeLibrary/${agent.role}/history/"
|
|
75
|
+
inbox: "KnowledgeLibrary/${agent.role}/inbox/"
|
|
76
|
+
outbox: "KnowledgeLibrary/${agent.role}/outbox/"
|
|
77
|
+
tech: "KnowledgeLibrary/${agent.role}/tech/"
|
|
78
|
+
control:
|
|
79
|
+
objectives: "KnowledgeLibrary/${agent.role}/control/objectives.txt"
|
|
80
|
+
decisions: "KnowledgeLibrary/${agent.role}/control/decisions.txt"
|
|
81
|
+
dependencies: "KnowledgeLibrary/${agent.role}/control/dependencies.txt"
|
|
82
|
+
index: "KnowledgeLibrary/${agent.role}/control/index.txt"
|
|
71
83
|
|
|
72
84
|
communication:
|
|
73
85
|
inbox: "KnowledgeLibrary/${agent.role}/inbox"
|
|
74
86
|
outbox: "KnowledgeLibrary/${agent.role}/outbox"
|
|
87
|
+
message_format: "YYYYMMDD_HHMM-[agent-name]-[topic].txt"
|
|
88
|
+
outbox_format: "YYYYMMDD_HHMM-to-[agent-name]-[topic].txt"
|
|
89
|
+
processed_dir: "inbox/processed/"
|
|
90
|
+
|
|
91
|
+
protocols:
|
|
92
|
+
startup:
|
|
93
|
+
first_session:
|
|
94
|
+
- "Read KnowledgeLibrary/context.txt"
|
|
95
|
+
- "Read shared documentation (${remote.documentation} primary, local fallback)"
|
|
96
|
+
- "Read KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
97
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for unprocessed messages"
|
|
98
|
+
- "Review control/objectives.txt and control/decisions.txt"
|
|
99
|
+
- "Read workflow definitions from KnowledgeLibrary/workflows.md"
|
|
100
|
+
subsequent:
|
|
101
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for new messages only"
|
|
102
|
+
completion:
|
|
103
|
+
- "Update KnowledgeLibrary/${agent.role}/context/current.txt with current state"
|
|
104
|
+
- "Archive previous context to history/ if significant changes (format: YYYYMMDD_HHMM-context.txt)"
|
|
105
|
+
- "Log key decisions to control/decisions.txt (format: [YYYY-MM-DD] Decision: [summary] | Rationale: [why])"
|
|
106
|
+
- "Store technical artifacts in tech/"
|
|
107
|
+
- "Mark inbox messages as processed (move to inbox/processed/)"
|
|
108
|
+
- "Send completion summary to engineering-manager inbox"
|
|
@@ -60,13 +60,47 @@ behaviors:
|
|
|
60
60
|
workflow: ticket-implementation
|
|
61
61
|
quality_gates: "${config.quality_gates}"
|
|
62
62
|
|
|
63
|
-
|
|
63
|
+
knowledge_library:
|
|
64
64
|
shared:
|
|
65
|
-
|
|
66
|
-
|
|
65
|
+
context: "KnowledgeLibrary/context.txt"
|
|
66
|
+
architecture: "KnowledgeLibrary/architecture.txt"
|
|
67
|
+
prd: "KnowledgeLibrary/prd.txt"
|
|
68
|
+
remote:
|
|
69
|
+
- "${remote.documentation}"
|
|
67
70
|
personal:
|
|
68
|
-
|
|
71
|
+
context: "KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
72
|
+
history: "KnowledgeLibrary/${agent.role}/history/"
|
|
73
|
+
inbox: "KnowledgeLibrary/${agent.role}/inbox/"
|
|
74
|
+
outbox: "KnowledgeLibrary/${agent.role}/outbox/"
|
|
75
|
+
tech: "KnowledgeLibrary/${agent.role}/tech/"
|
|
76
|
+
control:
|
|
77
|
+
objectives: "KnowledgeLibrary/${agent.role}/control/objectives.txt"
|
|
78
|
+
decisions: "KnowledgeLibrary/${agent.role}/control/decisions.txt"
|
|
79
|
+
dependencies: "KnowledgeLibrary/${agent.role}/control/dependencies.txt"
|
|
80
|
+
index: "KnowledgeLibrary/${agent.role}/control/index.txt"
|
|
69
81
|
|
|
70
82
|
communication:
|
|
71
83
|
inbox: "KnowledgeLibrary/${agent.role}/inbox"
|
|
72
84
|
outbox: "KnowledgeLibrary/${agent.role}/outbox"
|
|
85
|
+
message_format: "YYYYMMDD_HHMM-[agent-name]-[topic].txt"
|
|
86
|
+
outbox_format: "YYYYMMDD_HHMM-to-[agent-name]-[topic].txt"
|
|
87
|
+
processed_dir: "inbox/processed/"
|
|
88
|
+
|
|
89
|
+
protocols:
|
|
90
|
+
startup:
|
|
91
|
+
first_session:
|
|
92
|
+
- "Read KnowledgeLibrary/context.txt"
|
|
93
|
+
- "Read shared documentation (${remote.documentation} primary, local fallback)"
|
|
94
|
+
- "Read KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
95
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for unprocessed messages"
|
|
96
|
+
- "Review control/objectives.txt and control/decisions.txt"
|
|
97
|
+
- "Read workflow definitions from KnowledgeLibrary/workflows.md"
|
|
98
|
+
subsequent:
|
|
99
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for new messages only"
|
|
100
|
+
completion:
|
|
101
|
+
- "Update KnowledgeLibrary/${agent.role}/context/current.txt with current state"
|
|
102
|
+
- "Archive previous context to history/ if significant changes (format: YYYYMMDD_HHMM-context.txt)"
|
|
103
|
+
- "Log key decisions to control/decisions.txt (format: [YYYY-MM-DD] Decision: [summary] | Rationale: [why])"
|
|
104
|
+
- "Store technical artifacts in tech/"
|
|
105
|
+
- "Mark inbox messages as processed (move to inbox/processed/)"
|
|
106
|
+
- "Send completion summary to engineering-manager inbox"
|
|
@@ -58,13 +58,47 @@ behaviors:
|
|
|
58
58
|
workflow: planning-estimation
|
|
59
59
|
quality_gates: "${config.quality_gates}"
|
|
60
60
|
|
|
61
|
-
|
|
61
|
+
knowledge_library:
|
|
62
62
|
shared:
|
|
63
|
-
|
|
64
|
-
|
|
63
|
+
context: "KnowledgeLibrary/context.txt"
|
|
64
|
+
architecture: "KnowledgeLibrary/architecture.txt"
|
|
65
|
+
prd: "KnowledgeLibrary/prd.txt"
|
|
66
|
+
remote:
|
|
67
|
+
- "${remote.documentation}"
|
|
65
68
|
personal:
|
|
66
|
-
|
|
69
|
+
context: "KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
70
|
+
history: "KnowledgeLibrary/${agent.role}/history/"
|
|
71
|
+
inbox: "KnowledgeLibrary/${agent.role}/inbox/"
|
|
72
|
+
outbox: "KnowledgeLibrary/${agent.role}/outbox/"
|
|
73
|
+
tech: "KnowledgeLibrary/${agent.role}/tech/"
|
|
74
|
+
control:
|
|
75
|
+
objectives: "KnowledgeLibrary/${agent.role}/control/objectives.txt"
|
|
76
|
+
decisions: "KnowledgeLibrary/${agent.role}/control/decisions.txt"
|
|
77
|
+
dependencies: "KnowledgeLibrary/${agent.role}/control/dependencies.txt"
|
|
78
|
+
index: "KnowledgeLibrary/${agent.role}/control/index.txt"
|
|
67
79
|
|
|
68
80
|
communication:
|
|
69
81
|
inbox: "KnowledgeLibrary/${agent.role}/inbox"
|
|
70
82
|
outbox: "KnowledgeLibrary/${agent.role}/outbox"
|
|
83
|
+
message_format: "YYYYMMDD_HHMM-[agent-name]-[topic].txt"
|
|
84
|
+
outbox_format: "YYYYMMDD_HHMM-to-[agent-name]-[topic].txt"
|
|
85
|
+
processed_dir: "inbox/processed/"
|
|
86
|
+
|
|
87
|
+
protocols:
|
|
88
|
+
startup:
|
|
89
|
+
first_session:
|
|
90
|
+
- "Read KnowledgeLibrary/context.txt"
|
|
91
|
+
- "Read shared documentation (${remote.documentation} primary, local fallback)"
|
|
92
|
+
- "Read KnowledgeLibrary/${agent.role}/context/current.txt"
|
|
93
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for unprocessed messages"
|
|
94
|
+
- "Review control/objectives.txt and control/decisions.txt"
|
|
95
|
+
- "Read workflow definitions from KnowledgeLibrary/workflows.md"
|
|
96
|
+
subsequent:
|
|
97
|
+
- "Check KnowledgeLibrary/${agent.role}/inbox/ for new messages only"
|
|
98
|
+
completion:
|
|
99
|
+
- "Update KnowledgeLibrary/context.txt with current state, priorities, blockers"
|
|
100
|
+
- "Update KnowledgeLibrary/${agent.role}/context/current.txt with current state"
|
|
101
|
+
- "Archive previous context to history/ if significant changes (format: YYYYMMDD_HHMM-context.txt)"
|
|
102
|
+
- "Log key decisions to control/decisions.txt (format: [YYYY-MM-DD] Decision: [summary] | Rationale: [why])"
|
|
103
|
+
- "Mark inbox messages as processed (move to inbox/processed/)"
|
|
104
|
+
- "Delegate work to agent inboxes if applicable"
|