@atlashub/smartstack-cli 1.23.0 → 1.25.0
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/.documentation/agents.html +3 -3
- package/.documentation/apex.html +18 -18
- package/.documentation/business-analyse.html +38 -38
- package/.documentation/cli-commands.html +1 -1
- package/.documentation/commands.html +29 -29
- package/.documentation/efcore.html +246 -91
- package/.documentation/gitflow.html +1 -1
- package/.documentation/hooks.html +1 -1
- package/.documentation/index.html +8 -8
- package/.documentation/init.html +1 -1
- package/.documentation/installation.html +9 -9
- package/.documentation/ralph-loop.html +2 -2
- package/.documentation/test-web.html +2 -2
- package/dist/index.js +13 -0
- package/dist/index.js.map +1 -1
- package/package.json +1 -1
- package/templates/skills/check-version/SKILL.md +183 -0
- package/templates/skills/debug/SKILL.md +161 -0
- package/templates/skills/explore/SKILL.md +96 -0
- package/templates/skills/quick-search/SKILL.md +87 -0
- package/templates/skills/refactor/SKILL.md +219 -0
- package/templates/skills/review-code/SKILL.md +72 -44
- package/templates/skills/review-code/references/smartstack-conventions.md +93 -33
- package/templates/skills/ui-components/responsive-guidelines.md +278 -0
- package/templates/skills/utils/SKILL.md +37 -0
- package/templates/{commands/utils → skills/utils/subcommands}/test-web-config.md +35 -43
- package/templates/{commands/utils → skills/utils/subcommands}/test-web.md +25 -53
- package/templates/{commands/validate.md → skills/validate/SKILL.md} +80 -139
- package/templates/commands/check-version.md +0 -267
- package/templates/commands/debug.md +0 -95
- package/templates/commands/efcore/_env-check.md +0 -153
- package/templates/commands/efcore/_shared.md +0 -352
- package/templates/commands/efcore/conflicts.md +0 -90
- package/templates/commands/efcore/db-deploy.md +0 -109
- package/templates/commands/efcore/db-reset.md +0 -180
- package/templates/commands/efcore/db-seed.md +0 -103
- package/templates/commands/efcore/db-status.md +0 -102
- package/templates/commands/efcore/migration.md +0 -186
- package/templates/commands/efcore/rebase-snapshot.md +0 -172
- package/templates/commands/efcore/scan.md +0 -94
- package/templates/commands/efcore/squash.md +0 -329
- package/templates/commands/efcore.md +0 -96
- package/templates/commands/explore.md +0 -45
- package/templates/commands/quick-search.md +0 -72
- package/templates/commands/refactor.md +0 -164
- /package/templates/{commands → skills}/_resources/formatting-guide.md +0 -0
|
@@ -1,45 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
description: Deep codebase exploration to answer specific questions
|
|
3
|
-
argument-hint: <question>
|
|
4
|
-
---
|
|
5
|
-
|
|
6
|
-
You are a codebase exploration specialist. Answer questions through systematic investigation.
|
|
7
|
-
|
|
8
|
-
**You need to always ULTRA THINK.**
|
|
9
|
-
|
|
10
|
-
## Workflow
|
|
11
|
-
|
|
12
|
-
1. **PARSE QUESTION**: Understand what to investigate
|
|
13
|
-
- Extract key terms and concepts from question
|
|
14
|
-
- Identify file types, patterns, or areas to search
|
|
15
|
-
- Determine if web research is needed
|
|
16
|
-
|
|
17
|
-
2. **SEARCH CODEBASE**: Launch parallel exploration
|
|
18
|
-
- Use `explore-codebase` agents for code patterns
|
|
19
|
-
- Use `explore-docs` agents for library/framework specifics
|
|
20
|
-
- Use `websearch` agents if external context needed
|
|
21
|
-
- **CRITICAL**: Launch agents in parallel for speed
|
|
22
|
-
- Search for: implementations, configurations, examples, tests
|
|
23
|
-
|
|
24
|
-
3. **ANALYZE FINDINGS**: Synthesize discovered information
|
|
25
|
-
- Read relevant files found by agents
|
|
26
|
-
- Trace relationships between files
|
|
27
|
-
- Identify patterns and conventions
|
|
28
|
-
- Note file paths with line numbers (e.g., `src/app.ts:42`)
|
|
29
|
-
|
|
30
|
-
4. **ANSWER QUESTION**: Provide comprehensive response
|
|
31
|
-
- Direct answer to the question
|
|
32
|
-
- Supporting evidence with file references
|
|
33
|
-
- Code examples if relevant
|
|
34
|
-
- Architectural context when useful
|
|
35
|
-
|
|
36
|
-
## Execution Rules
|
|
37
|
-
|
|
38
|
-
- **PARALLEL SEARCH**: Launch multiple agents simultaneously
|
|
39
|
-
- **CITE SOURCES**: Always reference file paths and line numbers
|
|
40
|
-
- **STAY FOCUSED**: Only explore what's needed to answer the question
|
|
41
|
-
- **BE THOROUGH**: Don't stop at first match - gather complete context
|
|
42
|
-
|
|
43
|
-
## Priority
|
|
44
|
-
|
|
45
|
-
Accuracy > Speed > Brevity. Provide complete answers with evidence.
|
|
@@ -1,72 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
description: Lightning-fast search to answer specific questions - optimized for speed
|
|
3
|
-
argument-hint: <question>
|
|
4
|
-
allowed-tools: Grep, Glob, Read
|
|
5
|
-
model: haiku
|
|
6
|
-
---
|
|
7
|
-
|
|
8
|
-
You are a rapid search specialist. Answer questions at maximum speed using direct search tools.
|
|
9
|
-
|
|
10
|
-
## Workflow
|
|
11
|
-
|
|
12
|
-
1. **IDENTIFY**: Parse the question
|
|
13
|
-
|
|
14
|
-
- Extract key search terms
|
|
15
|
-
- Determine target file types or patterns
|
|
16
|
-
- **CRITICAL**: Be surgical - know exactly what to search
|
|
17
|
-
|
|
18
|
-
2. **SEARCH**: Direct tool usage (NO agents)
|
|
19
|
-
|
|
20
|
-
- Use `Grep` for code content search with specific patterns
|
|
21
|
-
- Use `Glob` for file name/path patterns
|
|
22
|
-
- Launch searches **in parallel** when possible
|
|
23
|
-
- **SPEED RULE**: Max 2-3 search iterations total
|
|
24
|
-
|
|
25
|
-
3. **READ**: Targeted file reading
|
|
26
|
-
|
|
27
|
-
- `Read` only the most relevant files found
|
|
28
|
-
- **CRITICAL**: Max 3-5 files - be selective
|
|
29
|
-
- Scan for the specific answer needed
|
|
30
|
-
|
|
31
|
-
4. **ANSWER**: Direct response
|
|
32
|
-
- Immediate answer to the question
|
|
33
|
-
- Include file references with line numbers (`file.ts:42`)
|
|
34
|
-
- **NO**: Long explanations or architectural context
|
|
35
|
-
- **YES**: Concise answer with evidence
|
|
36
|
-
|
|
37
|
-
## Execution Rules
|
|
38
|
-
|
|
39
|
-
- **SPEED FIRST**: Answer in under 30 seconds
|
|
40
|
-
- **NO AGENTS**: Use direct tools only (Grep, Glob, Read)
|
|
41
|
-
- **PARALLEL SEARCH**: Run independent searches simultaneously
|
|
42
|
-
- **MINIMAL READING**: Read only what's absolutely necessary
|
|
43
|
-
- **NO DEEP ANALYSIS**: Surface-level answers with citations
|
|
44
|
-
- **STOP EARLY**: Once you have the answer, respond immediately
|
|
45
|
-
|
|
46
|
-
## Search Patterns
|
|
47
|
-
|
|
48
|
-
**Finding implementations**:
|
|
49
|
-
|
|
50
|
-
```
|
|
51
|
-
Grep: pattern="class FooBar" or "function fooBar"
|
|
52
|
-
```
|
|
53
|
-
|
|
54
|
-
**Finding configs**:
|
|
55
|
-
|
|
56
|
-
```
|
|
57
|
-
Glob: pattern="**/config.{js,ts,json}"
|
|
58
|
-
```
|
|
59
|
-
|
|
60
|
-
**Finding imports/usage**:
|
|
61
|
-
|
|
62
|
-
```
|
|
63
|
-
Grep: pattern="from ['\"].*moduleName['\"]"
|
|
64
|
-
```
|
|
65
|
-
|
|
66
|
-
## Priority
|
|
67
|
-
|
|
68
|
-
Speed > Completeness. Fast answers beat perfect answers.
|
|
69
|
-
|
|
70
|
-
---
|
|
71
|
-
|
|
72
|
-
User: $ARGUMENTS
|
|
@@ -1,164 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
description: Refactor code by finding files, grouping them, and launching parallel Snipper agents
|
|
3
|
-
argument-hint: <search-pattern-or-description>
|
|
4
|
-
---
|
|
5
|
-
|
|
6
|
-
<objective>
|
|
7
|
-
Refactor code matching #$ARGUMENTS across the codebase using parallel Snipper agents for maximum speed.
|
|
8
|
-
|
|
9
|
-
This command finds all relevant files, creates ONE instruction file, then launches Snipper agents in parallel with batches of max 3 files each.
|
|
10
|
-
</objective>
|
|
11
|
-
|
|
12
|
-
<process>
|
|
13
|
-
|
|
14
|
-
## Phase 1: Discovery
|
|
15
|
-
|
|
16
|
-
1. **Parse the refactor request**: Understand what #$ARGUMENTS means
|
|
17
|
-
- Could be: method name, component name, pattern, code smell, etc.
|
|
18
|
-
- Identify the search strategy (Grep for code patterns, Glob for file patterns)
|
|
19
|
-
|
|
20
|
-
2. **Find all affected files**:
|
|
21
|
-
- Use Grep to search for the pattern in the codebase
|
|
22
|
-
- Use Glob if searching by file name patterns
|
|
23
|
-
- Exclude node_modules, .git, dist, build directories
|
|
24
|
-
- List all files that need refactoring
|
|
25
|
-
|
|
26
|
-
3. **Analyze scope**:
|
|
27
|
-
- Count total files found
|
|
28
|
-
- If more than 15 files, ask user to confirm or narrow scope
|
|
29
|
-
- Show preview of files to refactor
|
|
30
|
-
|
|
31
|
-
## Phase 2: Create Instructions
|
|
32
|
-
|
|
33
|
-
4. **Create task folder**:
|
|
34
|
-
- Generate unique ID: `refactor-{timestamp}`
|
|
35
|
-
- Create folder: `.claude/tasks/refactor-{timestamp}/`
|
|
36
|
-
|
|
37
|
-
5. **Create ONE instruction file**:
|
|
38
|
-
Create `.claude/tasks/refactor-{id}/instructions.md` with precise, adaptive instructions:
|
|
39
|
-
|
|
40
|
-
```markdown
|
|
41
|
-
# Refactor Instructions
|
|
42
|
-
|
|
43
|
-
## Objective
|
|
44
|
-
{Clear description of what to refactor based on #$ARGUMENTS}
|
|
45
|
-
|
|
46
|
-
## What to Change
|
|
47
|
-
{Specific patterns/code to find and modify}
|
|
48
|
-
|
|
49
|
-
## How to Change
|
|
50
|
-
{Step-by-step transformation rules}
|
|
51
|
-
|
|
52
|
-
## Rules
|
|
53
|
-
- Follow existing codebase patterns
|
|
54
|
-
- No unnecessary comments
|
|
55
|
-
- Preserve functionality
|
|
56
|
-
- Only modify what's necessary for this refactor
|
|
57
|
-
```
|
|
58
|
-
|
|
59
|
-
**IMPORTANT**: Make instructions adaptive - they should work for ANY file in the list.
|
|
60
|
-
|
|
61
|
-
## Phase 3: Group and Execute
|
|
62
|
-
|
|
63
|
-
6. **Group files into batches**:
|
|
64
|
-
- Maximum 3 files per batch
|
|
65
|
-
- Group by related functionality when possible
|
|
66
|
-
|
|
67
|
-
7. **Launch Snipper agents in parallel**:
|
|
68
|
-
For EACH batch, use Task tool with subagent_type='Snipper':
|
|
69
|
-
|
|
70
|
-
```
|
|
71
|
-
Using the instructions in .claude/tasks/refactor-{id}/instructions.md, refactor these files:
|
|
72
|
-
- {file_1}
|
|
73
|
-
- {file_2}
|
|
74
|
-
- {file_3}
|
|
75
|
-
```
|
|
76
|
-
|
|
77
|
-
**CRITICAL**: Launch ALL batches in a SINGLE message with multiple Task calls.
|
|
78
|
-
|
|
79
|
-
8. **Wait for completion**:
|
|
80
|
-
- All Snipper agents run in parallel
|
|
81
|
-
- Collect results from each
|
|
82
|
-
|
|
83
|
-
## Phase 4: Verification
|
|
84
|
-
|
|
85
|
-
9. **Validate changes**:
|
|
86
|
-
- Run `pnpm lint` or `npm run lint` to check for errors
|
|
87
|
-
- Run `pnpm tsc` or `npx tsc` if TypeScript project
|
|
88
|
-
- Report any failures
|
|
89
|
-
|
|
90
|
-
10. **Summary report**:
|
|
91
|
-
- List all files modified
|
|
92
|
-
- Show any errors encountered
|
|
93
|
-
- Provide next steps if needed
|
|
94
|
-
|
|
95
|
-
</process>
|
|
96
|
-
|
|
97
|
-
<instructions_template>
|
|
98
|
-
Create ONE file at `.claude/tasks/refactor-{id}/instructions.md`:
|
|
99
|
-
|
|
100
|
-
```markdown
|
|
101
|
-
# Refactor: {title}
|
|
102
|
-
|
|
103
|
-
## Objective
|
|
104
|
-
{What needs to be refactored - derived from #$ARGUMENTS}
|
|
105
|
-
|
|
106
|
-
## Pattern to Find
|
|
107
|
-
{Exact code pattern, method name, or structure to locate}
|
|
108
|
-
|
|
109
|
-
## Transformation
|
|
110
|
-
{How to transform the found pattern - be specific and adaptive}
|
|
111
|
-
|
|
112
|
-
## Examples
|
|
113
|
-
Before:
|
|
114
|
-
```
|
|
115
|
-
{example of current code}
|
|
116
|
-
```
|
|
117
|
-
|
|
118
|
-
After:
|
|
119
|
-
```
|
|
120
|
-
{example of refactored code}
|
|
121
|
-
```
|
|
122
|
-
|
|
123
|
-
## Constraints
|
|
124
|
-
- Only modify code matching the pattern
|
|
125
|
-
- Preserve all existing functionality
|
|
126
|
-
- Follow codebase conventions
|
|
127
|
-
- No comments unless necessary
|
|
128
|
-
```
|
|
129
|
-
</instructions_template>
|
|
130
|
-
|
|
131
|
-
<snipper_prompt_template>
|
|
132
|
-
For each batch, call Snipper with:
|
|
133
|
-
|
|
134
|
-
```
|
|
135
|
-
Using the instructions in .claude/tasks/refactor-{id}/instructions.md, refactor these files:
|
|
136
|
-
- {file_path_1}
|
|
137
|
-
- {file_path_2}
|
|
138
|
-
- {file_path_3}
|
|
139
|
-
|
|
140
|
-
Read the instructions file first, then apply the refactor to each file.
|
|
141
|
-
```
|
|
142
|
-
</snipper_prompt_template>
|
|
143
|
-
|
|
144
|
-
<success_criteria>
|
|
145
|
-
- All target files identified
|
|
146
|
-
- ONE instruction file created in `.claude/tasks/refactor-{id}/instructions.md`
|
|
147
|
-
- Snipper agents launched in parallel (max 3 files per agent)
|
|
148
|
-
- All batches completed successfully
|
|
149
|
-
- Lint/type checks pass
|
|
150
|
-
- Summary provided to user
|
|
151
|
-
</success_criteria>
|
|
152
|
-
|
|
153
|
-
<example>
|
|
154
|
-
User: `/refactor rename getUserData to fetchUserProfile`
|
|
155
|
-
|
|
156
|
-
1. Grep finds 12 files containing "getUserData"
|
|
157
|
-
2. Creates `.claude/tasks/refactor-1702489200/`
|
|
158
|
-
3. Creates ONE file: `instructions.md` with rename rules
|
|
159
|
-
4. Groups into 4 batches of 3 files
|
|
160
|
-
5. Launches 4 Snipper agents in parallel, each with:
|
|
161
|
-
"Using instructions in .../instructions.md, refactor: file1, file2, file3"
|
|
162
|
-
6. Waits for completion, runs lint
|
|
163
|
-
7. Reports: "Refactored 12 files, renamed getUserData to fetchUserProfile"
|
|
164
|
-
</example>
|
|
File without changes
|