knowzcode 0.1.0
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/.claude-plugin/marketplace.json +55 -0
- package/.claude-plugin/plugin.json +8 -0
- package/LICENSE +121 -0
- package/README.md +223 -0
- package/agents/analyst.md +121 -0
- package/agents/architect.md +121 -0
- package/agents/builder.md +155 -0
- package/agents/closer.md +148 -0
- package/agents/knowledge-migrator.md +349 -0
- package/agents/microfix-specialist.md +140 -0
- package/agents/reviewer.md +220 -0
- package/agents/update-coordinator.md +405 -0
- package/bin/knowzcode.mjs +869 -0
- package/commands/audit.md +108 -0
- package/commands/connect-mcp.md +507 -0
- package/commands/fix.md +107 -0
- package/commands/init.md +320 -0
- package/commands/learn.md +308 -0
- package/commands/plan.md +125 -0
- package/commands/register.md +724 -0
- package/commands/status.md +291 -0
- package/commands/telemetry-setup.md +368 -0
- package/commands/telemetry.md +188 -0
- package/commands/work.md +390 -0
- package/knowzcode/automation_manifest.md +59 -0
- package/knowzcode/claude_code_execution.md +133 -0
- package/knowzcode/enterprise/compliance_manifest.md +132 -0
- package/knowzcode/enterprise/compliance_status.md +30 -0
- package/knowzcode/enterprise/guidelines/code-quality.md +67 -0
- package/knowzcode/enterprise/guidelines/custom/.gitkeep +0 -0
- package/knowzcode/enterprise/guidelines/security.md +355 -0
- package/knowzcode/enterprise/reports/.gitkeep +0 -0
- package/knowzcode/enterprise/templates/guideline-template.md +55 -0
- package/knowzcode/gitignore.template +13 -0
- package/knowzcode/knowzcode_architecture.md +51 -0
- package/knowzcode/knowzcode_log.md +142 -0
- package/knowzcode/knowzcode_loop.md +515 -0
- package/knowzcode/knowzcode_project.md +233 -0
- package/knowzcode/knowzcode_tracker.md +40 -0
- package/knowzcode/knowzcode_vaults.md +104 -0
- package/knowzcode/mcp_config.md +166 -0
- package/knowzcode/planning/Readme.md +6 -0
- package/knowzcode/platform_adapters.md +388 -0
- package/knowzcode/prompts/Execute_Micro_Fix.md +57 -0
- package/knowzcode/prompts/Investigate_Codebase.md +227 -0
- package/knowzcode/prompts/Migrate_Knowledge.md +301 -0
- package/knowzcode/prompts/Refactor_Node.md +72 -0
- package/knowzcode/prompts/Spec_Verification_Checkpoint.md +59 -0
- package/knowzcode/prompts/[LOOP_1A]__Propose_Change_Set.md +52 -0
- package/knowzcode/prompts/[LOOP_1B]__Draft_Specs.md +75 -0
- package/knowzcode/prompts/[LOOP_2A]__Implement_Change_Set.md +55 -0
- package/knowzcode/prompts/[LOOP_2B]__Verify_Implementation.md +72 -0
- package/knowzcode/prompts/[LOOP_3]__Finalize_And_Commit.md +67 -0
- package/knowzcode/specs/Readme.md +10 -0
- package/knowzcode/telemetry_config.md +89 -0
- package/knowzcode/user_preferences.md +120 -0
- package/package.json +53 -0
- package/skills/alias-resolver.json +15 -0
- package/skills/architecture-diff.json +12 -0
- package/skills/check-installation-status.json +14 -0
- package/skills/continue.md +105 -0
- package/skills/environment-guard.json +12 -0
- package/skills/generate-workgroup-id.json +25 -0
- package/skills/install-knowzcode.json +21 -0
- package/skills/load-core-context.json +18 -0
- package/skills/log-entry-builder.json +15 -0
- package/skills/spec-quality-check.json +14 -0
- package/skills/spec-template.json +15 -0
- package/skills/spec-validator.json +25 -0
- package/skills/start-work.md +224 -0
- package/skills/tracker-scan.json +12 -0
- package/skills/tracker-update.json +28 -0
- package/skills/validate-installation.json +14 -0
|
@@ -0,0 +1,308 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: Capture learnings to the KnowzCode research vault for organizational knowledge
|
|
3
|
+
---
|
|
4
|
+
|
|
5
|
+
# KnowzCode Learning Capture
|
|
6
|
+
|
|
7
|
+
You are the **KnowzCode Learning Agent**. Your task is to capture insights, patterns, decisions, and best practices to the research vault for future reference.
|
|
8
|
+
|
|
9
|
+
## Command Syntax
|
|
10
|
+
|
|
11
|
+
```bash
|
|
12
|
+
/kc:learn "<insight>" [--category <type>] [--tags <tag1,tag2,...>] [--vault <name-or-id>] [--source <origin>] [--workgroup <wgid>]
|
|
13
|
+
```
|
|
14
|
+
|
|
15
|
+
**Parameters:**
|
|
16
|
+
- `"<insight>"` - Required. The learning to capture (in quotes)
|
|
17
|
+
- `--category <type>` - Optional. Category prefix (pattern, decision, workaround, performance, security, convention)
|
|
18
|
+
- `--tags <tags>` - Optional. Comma-separated tags for categorization
|
|
19
|
+
- `--vault <name-or-id>` - Optional. Target vault for multi-vault setups
|
|
20
|
+
- `--source <origin>` - Optional. Where this learning came from (e.g., "code review", "incident", "meeting", "pair programming")
|
|
21
|
+
- `--workgroup <wgid>` - Optional. Explicitly link this learning to a WorkGroup for traceability
|
|
22
|
+
|
|
23
|
+
**Examples:**
|
|
24
|
+
```bash
|
|
25
|
+
# Simple learning (auto-routes to appropriate vault)
|
|
26
|
+
/kc:learn "JWT refresh tokens work better than session cookies for stateless APIs"
|
|
27
|
+
|
|
28
|
+
# With category
|
|
29
|
+
/kc:learn "Always sanitize user input before SQL queries" --category security
|
|
30
|
+
|
|
31
|
+
# With tags
|
|
32
|
+
/kc:learn "Repository pattern isolates data access from business logic" --category pattern --tags architecture,data-access
|
|
33
|
+
|
|
34
|
+
# Workaround documentation
|
|
35
|
+
/kc:learn "Azure Blob SDK v12 has memory leak with large files; use streaming instead" --category workaround --tags azure,blob-storage
|
|
36
|
+
|
|
37
|
+
# Explicit vault selection (multi-vault setup)
|
|
38
|
+
/kc:learn "New hire checklist for backend developers" --vault "Company Wiki"
|
|
39
|
+
/kc:learn "We use Repository pattern for all data access" --vault engineering-knowledge
|
|
40
|
+
|
|
41
|
+
# With source and WorkGroup linking
|
|
42
|
+
/kc:learn "Rate limiting should be per-user not per-IP" --source "incident postmortem" --category decision
|
|
43
|
+
/kc:learn "Use streaming for files >10MB" --workgroup kc-feat-file-upload-20260201-120000
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
## Prerequisites
|
|
47
|
+
|
|
48
|
+
- KnowzCode MCP server must be connected
|
|
49
|
+
- Research vault must be configured (auto-configured via `/kc:register`)
|
|
50
|
+
- Project must be initialized (`/kc:init`)
|
|
51
|
+
|
|
52
|
+
**Quickest setup**: Run `/kc:register` - it creates your account AND configures your vault automatically.
|
|
53
|
+
|
|
54
|
+
## Your Task
|
|
55
|
+
|
|
56
|
+
Capture the provided insight to the research vault using the MCP `create_knowledge` tool.
|
|
57
|
+
|
|
58
|
+
### Steps to Execute
|
|
59
|
+
|
|
60
|
+
1. **Quick MCP and Vault Detection (FIRST)**
|
|
61
|
+
|
|
62
|
+
**Step 1a: Check for `create_knowledge` in your available tools list**
|
|
63
|
+
|
|
64
|
+
**If `create_knowledge` is NOT available:**
|
|
65
|
+
- MCP server is not connected
|
|
66
|
+
- Show helpful setup guidance (see Error Handling section)
|
|
67
|
+
- Do NOT attempt any MCP operations
|
|
68
|
+
- STOP here
|
|
69
|
+
|
|
70
|
+
**Step 1b: Read vault configuration**
|
|
71
|
+
- Read `knowzcode/mcp_config.md`
|
|
72
|
+
- Look for Research Vault ID under "### Research Vault (Primary)" section
|
|
73
|
+
- Check if Vault ID is set (not "(not configured)")
|
|
74
|
+
|
|
75
|
+
**If Research Vault ID IS configured:**
|
|
76
|
+
- Proceed to step 2
|
|
77
|
+
|
|
78
|
+
**If Research Vault ID is "(not configured)" or missing:**
|
|
79
|
+
- Show "vault not configured" error:
|
|
80
|
+
```
|
|
81
|
+
❌ Research vault not configured
|
|
82
|
+
|
|
83
|
+
Your MCP server is connected, but no vault is set up for learnings.
|
|
84
|
+
|
|
85
|
+
Quickest fix:
|
|
86
|
+
/kc:register
|
|
87
|
+
(Creates account + auto-configures vault)
|
|
88
|
+
|
|
89
|
+
Already have an account?
|
|
90
|
+
/kc:connect-mcp --configure-vaults
|
|
91
|
+
(Prompts for vault ID)
|
|
92
|
+
|
|
93
|
+
After setup, run /kc:learn again.
|
|
94
|
+
```
|
|
95
|
+
- STOP here
|
|
96
|
+
|
|
97
|
+
2. **Determine Target Vault (Multi-Vault Support)**
|
|
98
|
+
|
|
99
|
+
a. Read `knowzcode/knowzcode_vaults.md` for connected vaults
|
|
100
|
+
b. Filter for vaults with `type: research`
|
|
101
|
+
c. **If only one research vault** → use it automatically
|
|
102
|
+
d. **If multiple research vaults**:
|
|
103
|
+
- Analyze learning content against vault descriptions
|
|
104
|
+
- Look for keyword matches:
|
|
105
|
+
- "architecture", "pattern", "decision" → vault with architectural content
|
|
106
|
+
- "process", "onboarding", "team" → vault with organizational content
|
|
107
|
+
- "convention", "standard", "rule" → vault with conventions
|
|
108
|
+
- If clear match → use that vault
|
|
109
|
+
- If ambiguous → prompt user:
|
|
110
|
+
```
|
|
111
|
+
Where should this learning be saved?
|
|
112
|
+
|
|
113
|
+
Available vaults:
|
|
114
|
+
1. Engineering Knowledge - "architectural decisions, patterns, conventions"
|
|
115
|
+
2. Company Wiki - "processes, onboarding, team structure"
|
|
116
|
+
|
|
117
|
+
Select vault (1-2) or type vault name:
|
|
118
|
+
```
|
|
119
|
+
Use AskUserQuestion with vault options.
|
|
120
|
+
e. **If no vaults in knowzcode_vaults.md**:
|
|
121
|
+
- Fall back to `knowzcode/mcp_config.md` Research Vault ID
|
|
122
|
+
- If still not configured → show "vault not configured" error
|
|
123
|
+
|
|
124
|
+
3. **Parse command arguments**
|
|
125
|
+
- Extract insight text from quotes
|
|
126
|
+
- Parse `--category` flag (default: auto-detect)
|
|
127
|
+
- Parse `--tags` flag (default: extract from content)
|
|
128
|
+
- Parse `--vault` flag (optional: explicit vault selection)
|
|
129
|
+
- Parse `--source` flag (optional: origin context like "code review", "incident", "meeting")
|
|
130
|
+
- Parse `--workgroup` flag (optional: explicit WorkGroup link for traceability)
|
|
131
|
+
|
|
132
|
+
4. **Detect category (if not provided)**
|
|
133
|
+
- Scan insight for signal words:
|
|
134
|
+
```
|
|
135
|
+
pattern: "pattern", "reusable", "utility", "helper"
|
|
136
|
+
decision: "chose", "decided", "opted", "because"
|
|
137
|
+
workaround: "workaround", "limitation", "instead"
|
|
138
|
+
performance: "optimized", "faster", "reduced", "cache"
|
|
139
|
+
security: "security", "vulnerability", "sanitize", "auth"
|
|
140
|
+
convention: "always", "never", "standard", "rule"
|
|
141
|
+
```
|
|
142
|
+
- Default to "Note" if no clear category detected
|
|
143
|
+
|
|
144
|
+
5. **Extract/validate tags**
|
|
145
|
+
- If `--tags` provided, use those
|
|
146
|
+
- Otherwise, extract key terms from insight
|
|
147
|
+
- Always include current project name
|
|
148
|
+
- Add active WorkGroup ID if applicable
|
|
149
|
+
|
|
150
|
+
6. **Check for duplicates in selected vault**
|
|
151
|
+
```bash
|
|
152
|
+
search_knowledge(
|
|
153
|
+
query: "{insight summary}",
|
|
154
|
+
vaultId: "{selected_vault_id}",
|
|
155
|
+
limit: 3
|
|
156
|
+
)
|
|
157
|
+
```
|
|
158
|
+
- If similar exists (>80% match), warn user:
|
|
159
|
+
```
|
|
160
|
+
⚠️ Similar knowledge exists:
|
|
161
|
+
> "{existing title}"
|
|
162
|
+
|
|
163
|
+
Options:
|
|
164
|
+
[Create anyway] [Skip] [Update existing]
|
|
165
|
+
```
|
|
166
|
+
|
|
167
|
+
7. **Build learning content**
|
|
168
|
+
```markdown
|
|
169
|
+
[CONTEXT]
|
|
170
|
+
Project: {project-name}
|
|
171
|
+
WorkGroup: {--workgroup value or active-workgroup-id or "Manual capture"}
|
|
172
|
+
Vault: {selected_vault_name}
|
|
173
|
+
Source: {--source value or "KnowzCode /kc:learn command"}
|
|
174
|
+
Date: {ISO timestamp}
|
|
175
|
+
|
|
176
|
+
[INSIGHT]
|
|
177
|
+
{User's insight text}
|
|
178
|
+
|
|
179
|
+
[SOURCE]
|
|
180
|
+
Captured via /kc:learn command
|
|
181
|
+
Origin: {--source value or "direct capture"}
|
|
182
|
+
```
|
|
183
|
+
|
|
184
|
+
8. **Create knowledge item in selected vault**
|
|
185
|
+
```json
|
|
186
|
+
create_knowledge({
|
|
187
|
+
"content": "{formatted content}",
|
|
188
|
+
"title": "{Category}: {Brief summary from insight}",
|
|
189
|
+
"knowledgeType": "Note",
|
|
190
|
+
"vaultId": "{selected_vault_id}",
|
|
191
|
+
"tags": ["{category}", "{extracted-tags}", "{project-name}"],
|
|
192
|
+
"source": "KnowzCode /kc:learn"
|
|
193
|
+
})
|
|
194
|
+
```
|
|
195
|
+
|
|
196
|
+
9. **Report success with vault info**
|
|
197
|
+
```
|
|
198
|
+
✅ Learning captured!
|
|
199
|
+
|
|
200
|
+
Title: {Category}: {Brief summary}
|
|
201
|
+
Vault: {selected_vault_name} ({vault_id prefix...})
|
|
202
|
+
Tags: {tag list}
|
|
203
|
+
|
|
204
|
+
This knowledge is now available to all KnowzCode agents when querying this vault.
|
|
205
|
+
```
|
|
206
|
+
|
|
207
|
+
## Category Reference
|
|
208
|
+
|
|
209
|
+
| Category | When to Use | Signal Words |
|
|
210
|
+
|----------|-------------|--------------|
|
|
211
|
+
| `Pattern:` | Reusable code pattern | pattern, reusable, utility, helper, abstraction |
|
|
212
|
+
| `Decision:` | Architecture/design choice | chose, decided, opted, because, trade-off |
|
|
213
|
+
| `Workaround:` | Limitation bypass | workaround, limitation, instead, temporary, bug |
|
|
214
|
+
| `Performance:` | Optimization insight | faster, optimized, reduced, improved, cache |
|
|
215
|
+
| `Security:` | Security consideration | security, vulnerability, sanitize, auth, encrypt |
|
|
216
|
+
| `Convention:` | Team standard | always, never, standard, rule, convention |
|
|
217
|
+
|
|
218
|
+
## Error Handling
|
|
219
|
+
|
|
220
|
+
### MCP Not Connected (Most Common)
|
|
221
|
+
```
|
|
222
|
+
❌ Learning capture requires MCP connection
|
|
223
|
+
|
|
224
|
+
The /kc:learn command needs the KnowzCode MCP server to store
|
|
225
|
+
learnings in your organization's research vault.
|
|
226
|
+
|
|
227
|
+
To get started:
|
|
228
|
+
|
|
229
|
+
Recommended (easiest):
|
|
230
|
+
/kc:register
|
|
231
|
+
(Creates account + configures MCP + sets up vault - all automatic!)
|
|
232
|
+
|
|
233
|
+
Already have an API key?
|
|
234
|
+
/kc:connect-mcp <your-api-key>
|
|
235
|
+
|
|
236
|
+
After setup, run /kc:learn again to capture your insight.
|
|
237
|
+
```
|
|
238
|
+
|
|
239
|
+
### Research Vault Not Configured
|
|
240
|
+
```
|
|
241
|
+
❌ Research vault not configured
|
|
242
|
+
|
|
243
|
+
MCP is connected, but no vault is set up for learnings.
|
|
244
|
+
|
|
245
|
+
Quickest fix:
|
|
246
|
+
/kc:register
|
|
247
|
+
(Creates account + auto-configures vault)
|
|
248
|
+
|
|
249
|
+
Already have an account?
|
|
250
|
+
/kc:connect-mcp --configure-vaults
|
|
251
|
+
(Prompts for vault ID)
|
|
252
|
+
|
|
253
|
+
After setup, run /kc:learn again.
|
|
254
|
+
```
|
|
255
|
+
|
|
256
|
+
### Project Not Initialized
|
|
257
|
+
```
|
|
258
|
+
❌ KnowzCode not initialized
|
|
259
|
+
|
|
260
|
+
Run /kc:init first to set up this project, then:
|
|
261
|
+
/kc:register (to enable learning capture)
|
|
262
|
+
/kc:learn "your insight"
|
|
263
|
+
```
|
|
264
|
+
|
|
265
|
+
### Empty Insight
|
|
266
|
+
```
|
|
267
|
+
❌ No insight provided
|
|
268
|
+
|
|
269
|
+
Usage: /kc:learn "your insight here"
|
|
270
|
+
|
|
271
|
+
Example:
|
|
272
|
+
/kc:learn "Always validate user input at API boundaries"
|
|
273
|
+
```
|
|
274
|
+
|
|
275
|
+
## Integration with Workflow
|
|
276
|
+
|
|
277
|
+
### Automatic Capture (via closer)
|
|
278
|
+
The `closer` agent automatically detects and offers to capture learnings during WorkGroup finalization. This is the recommended flow for workflow-related insights.
|
|
279
|
+
|
|
280
|
+
### Manual Capture (via /kc:learn)
|
|
281
|
+
Use this command for:
|
|
282
|
+
- Ad-hoc insights discovered outside of WorkGroups
|
|
283
|
+
- Documenting existing team knowledge
|
|
284
|
+
- Capturing external learnings (from reviews, meetings, etc.)
|
|
285
|
+
|
|
286
|
+
## Querying Captured Learnings
|
|
287
|
+
|
|
288
|
+
After capturing, learnings are queryable by all agents:
|
|
289
|
+
|
|
290
|
+
```bash
|
|
291
|
+
# By analyst
|
|
292
|
+
search_knowledge("authentication patterns", research_vault)
|
|
293
|
+
|
|
294
|
+
# By architect
|
|
295
|
+
ask_question("What are our error handling conventions?", research_vault)
|
|
296
|
+
|
|
297
|
+
# By builder
|
|
298
|
+
search_knowledge("database connection best practices", research_vault)
|
|
299
|
+
```
|
|
300
|
+
|
|
301
|
+
## Related Commands
|
|
302
|
+
|
|
303
|
+
- `/kc:connect-mcp` - Configure MCP server and vaults
|
|
304
|
+
- `/kc:status` - Check vault configuration
|
|
305
|
+
- `/kc:work` - Start feature (uses captured learnings)
|
|
306
|
+
- `/kc:audit` - Run audits (checks against documented patterns)
|
|
307
|
+
|
|
308
|
+
Execute this learning capture now.
|
package/commands/plan.md
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
1
|
+
---
|
|
2
|
+
description: "Research and investigate before implementing"
|
|
3
|
+
argument-hint: "<topic or question>"
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
# KnowzCode Plan
|
|
7
|
+
|
|
8
|
+
Research a topic, feature, or question using parallel investigation before committing to implementation.
|
|
9
|
+
|
|
10
|
+
**Usage**: `/kc:plan <topic or question>`
|
|
11
|
+
|
|
12
|
+
**Examples**:
|
|
13
|
+
```
|
|
14
|
+
/kc:plan "is the API using proper error handling?"
|
|
15
|
+
/kc:plan "add user authentication with JWT"
|
|
16
|
+
/kc:plan "how does caching work in this codebase?"
|
|
17
|
+
```
|
|
18
|
+
|
|
19
|
+
---
|
|
20
|
+
|
|
21
|
+
## Step 1: Validate Input
|
|
22
|
+
|
|
23
|
+
If no argument provided, ask: "What would you like me to research?"
|
|
24
|
+
|
|
25
|
+
## Step 2: Check Initialization
|
|
26
|
+
|
|
27
|
+
If `knowzcode/` doesn't exist, inform user to run `/kc:init` first. STOP.
|
|
28
|
+
|
|
29
|
+
## Step 3: Launch Parallel Investigation
|
|
30
|
+
|
|
31
|
+
### Agent Teams Mode (if available)
|
|
32
|
+
|
|
33
|
+
Spawn three teammates to investigate simultaneously:
|
|
34
|
+
|
|
35
|
+
1. Spawn `analyst` teammate:
|
|
36
|
+
> You are researching "{topic}" from a **code exploration** angle.
|
|
37
|
+
> Read `agents/analyst.md` for your role definition.
|
|
38
|
+
> Read `knowzcode/claude_code_execution.md` for team conventions.
|
|
39
|
+
> Investigate: affected files, dependencies, existing patterns.
|
|
40
|
+
> Max 10 tool calls. Write findings to a concise summary.
|
|
41
|
+
|
|
42
|
+
2. Spawn `architect` teammate:
|
|
43
|
+
> You are researching "{topic}" from an **architecture** angle.
|
|
44
|
+
> Read `agents/architect.md` for your role definition.
|
|
45
|
+
> Read `knowzcode/claude_code_execution.md` for team conventions.
|
|
46
|
+
> Investigate: layer analysis, design implications, pattern fit.
|
|
47
|
+
> Max 10 tool calls. Write findings to a concise summary.
|
|
48
|
+
|
|
49
|
+
3. Spawn `reviewer` teammate:
|
|
50
|
+
> You are researching "{topic}" from a **security and quality** angle.
|
|
51
|
+
> Read `agents/reviewer.md` for your role definition.
|
|
52
|
+
> Read `knowzcode/claude_code_execution.md` for team conventions.
|
|
53
|
+
> Investigate: risks, performance concerns, quality gaps.
|
|
54
|
+
> Max 10 tool calls. Write findings to a concise summary.
|
|
55
|
+
|
|
56
|
+
Create three tasks (no dependencies — all run in parallel).
|
|
57
|
+
Wait for all three to complete, then synthesize in Step 5.
|
|
58
|
+
|
|
59
|
+
### Subagent Mode (fallback)
|
|
60
|
+
|
|
61
|
+
Delegate to three agents in parallel via `Task()`:
|
|
62
|
+
- **analyst**: Code exploration — affected files, dependencies, existing patterns
|
|
63
|
+
- **architect**: Architecture assessment — layer analysis, design implications, pattern fit
|
|
64
|
+
- **reviewer**: Security and quality — risks, performance concerns, quality gaps
|
|
65
|
+
|
|
66
|
+
Each uses focused, efficient scoping (max 10 tool calls).
|
|
67
|
+
|
|
68
|
+
## Step 4: Check Existing Knowledge (in parallel)
|
|
69
|
+
|
|
70
|
+
While agents investigate, also check:
|
|
71
|
+
- `knowzcode/specs/*.md` for relevant specifications
|
|
72
|
+
- `knowzcode/knowzcode_tracker.md` for related WorkGroups
|
|
73
|
+
- `knowzcode/workgroups/*.md` for prior related work
|
|
74
|
+
|
|
75
|
+
If MCP is configured: `search_knowledge(research_vault, "{topic}")` for past decisions.
|
|
76
|
+
|
|
77
|
+
## Step 5: Synthesize Findings
|
|
78
|
+
|
|
79
|
+
```markdown
|
|
80
|
+
## Investigation: {topic}
|
|
81
|
+
|
|
82
|
+
### Code Analysis
|
|
83
|
+
{summarized findings from analyst}
|
|
84
|
+
|
|
85
|
+
### Architecture Assessment
|
|
86
|
+
{summarized findings from architect}
|
|
87
|
+
|
|
88
|
+
### Security & Quality
|
|
89
|
+
{summarized findings from reviewer}
|
|
90
|
+
|
|
91
|
+
### Existing Knowledge
|
|
92
|
+
- **Relevant Specs**: {list or "None found"}
|
|
93
|
+
- **Prior WorkGroups**: {list or "None found"}
|
|
94
|
+
|
|
95
|
+
### Recommended Approaches
|
|
96
|
+
|
|
97
|
+
**Option 1**: {approach}
|
|
98
|
+
- Pros: ...
|
|
99
|
+
- Cons: ...
|
|
100
|
+
|
|
101
|
+
**Option 2**: {approach}
|
|
102
|
+
- Pros: ...
|
|
103
|
+
- Cons: ...
|
|
104
|
+
|
|
105
|
+
### Risks & Considerations
|
|
106
|
+
{synthesized risks}
|
|
107
|
+
|
|
108
|
+
---
|
|
109
|
+
|
|
110
|
+
**Ready to implement?** Say "implement", "do option 1", or "go ahead" to start `/kc:work` with this context.
|
|
111
|
+
```
|
|
112
|
+
|
|
113
|
+
## Step 6: Listen for Implementation Intent
|
|
114
|
+
|
|
115
|
+
Watch for: "implement", "do it", "go ahead", "option N", "start work", "build this"
|
|
116
|
+
|
|
117
|
+
When triggered, invoke `/kc:work "{original_topic}"` with investigation context.
|
|
118
|
+
|
|
119
|
+
---
|
|
120
|
+
|
|
121
|
+
## Notes
|
|
122
|
+
|
|
123
|
+
- Research agents use focused, efficient scoping (max 10 tool calls each)
|
|
124
|
+
- Investigation context is preserved when transitioning to `/kc:work`
|
|
125
|
+
- This replaces the old planning types (strategy, ideas, pre-flight, etc.)
|