@simplysm/sd-claude 13.0.68 → 13.0.69
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/skills/sd-use/SKILL.md +26 -24
- package/package.json +1 -1
|
@@ -7,7 +7,8 @@ model: haiku
|
|
|
7
7
|
|
|
8
8
|
# sd-use - Auto Skill/Agent Router
|
|
9
9
|
|
|
10
|
-
Analyze user request from ARGUMENTS, select the best matching sd-\* skill or agent, explain why,
|
|
10
|
+
Analyze user request from ARGUMENTS, select the best matching sd-\* skill or agent, explain why,
|
|
11
|
+
then execute it.
|
|
11
12
|
|
|
12
13
|
## Execution Flow
|
|
13
14
|
|
|
@@ -20,37 +21,38 @@ Analyze user request from ARGUMENTS, select the best matching sd-\* skill or age
|
|
|
20
21
|
|
|
21
22
|
### Skills (execute via `Skill` tool)
|
|
22
23
|
|
|
23
|
-
| Skill | When to select
|
|
24
|
-
|
|
25
|
-
| `sd-brainstorm` | New feature, component, or behavior change — **creative work before implementation**
|
|
26
|
-
| `sd-debug` | Bug, test failure, unexpected behavior — **systematic root cause investigation**
|
|
27
|
-
| `sd-tdd` | Implementing a feature or fixing a bug — **before writing code**
|
|
28
|
-
| `sd-plan` | Multi-step task with spec/requirements — **planning before code**
|
|
29
|
-
| `sd-plan-dev` | Already have a plan — **executing implementation plan**
|
|
30
|
-
| `sd-explore` | Deep codebase analysis — tracing execution paths, architecture, dependencies
|
|
24
|
+
| Skill | When to select |
|
|
25
|
+
|----------------------|-------------------------------------------------------------------------------------------------------------------------------------------------|
|
|
26
|
+
| `sd-brainstorm` | New feature, component, or behavior change — **creative work before implementation** |
|
|
27
|
+
| `sd-debug` | Bug, test failure, unexpected behavior — **systematic root cause investigation** |
|
|
28
|
+
| `sd-tdd` | Implementing a feature or fixing a bug — **before writing code** |
|
|
29
|
+
| `sd-plan` | Multi-step task with spec/requirements — **planning before code** |
|
|
30
|
+
| `sd-plan-dev` | Already have a plan — **executing implementation plan** |
|
|
31
|
+
| `sd-explore` | Deep codebase analysis — tracing execution paths, architecture, dependencies |
|
|
31
32
|
| `sd-review` | **Large-scale** comprehensive review of an entire package or broad path — use only when user explicitly requests full/deep/comprehensive review |
|
|
32
|
-
| `sd-check` | Verify code — typecheck, lint, tests
|
|
33
|
-
| `sd-commit` | Create a git commit
|
|
34
|
-
| `sd-readme` | Update a package README.md
|
|
35
|
-
| `sd-discuss` | Evaluate code design decisions against industry standards and project conventions
|
|
36
|
-
| `sd-api-name-review` | Review public API naming consistency
|
|
37
|
-
| `sd-worktree` | Start new work in branch isolation
|
|
38
|
-
| `sd-skill` | Create or edit skills
|
|
39
|
-
| `sd-email-analyze` | Analyze, read, or summarize email files (`.eml` or `.msg`) — parsing and attachment extraction
|
|
33
|
+
| `sd-check` | Verify code — typecheck, lint, tests |
|
|
34
|
+
| `sd-commit` | Create a git commit |
|
|
35
|
+
| `sd-readme` | Update a package README.md |
|
|
36
|
+
| `sd-discuss` | Evaluate code design decisions against industry standards and project conventions |
|
|
37
|
+
| `sd-api-name-review` | Review public API naming consistency |
|
|
38
|
+
| `sd-worktree` | Start new work in branch isolation |
|
|
39
|
+
| `sd-skill` | Create or edit skills |
|
|
40
|
+
| `sd-email-analyze` | Analyze, read, or summarize email files (`.eml` or `.msg`) — parsing and attachment extraction |
|
|
40
41
|
|
|
41
42
|
### Agents (execute via `Task` tool with matching `subagent_type`)
|
|
42
43
|
|
|
43
|
-
| Agent
|
|
44
|
-
|
|
45
|
-
| `sd-code-reviewer`
|
|
46
|
-
| `sd-code-simplifier`
|
|
47
|
-
| `sd-api-reviewer`
|
|
48
|
-
| `sd-security-reviewer` | ORM SQL injection and input validation vulnerability review
|
|
44
|
+
| Agent | When to select |
|
|
45
|
+
|------------------------|------------------------------------------------------------------------------------------------------------------------------------|
|
|
46
|
+
| `sd-code-reviewer` | Quick/focused review — specific files, recent changes, bugs, security, quality issues. **Default choice for most review requests** |
|
|
47
|
+
| `sd-code-simplifier` | Simplify, clean up, improve code readability |
|
|
48
|
+
| `sd-api-reviewer` | Review library public API for DX quality |
|
|
49
|
+
| `sd-security-reviewer` | ORM SQL injection and input validation vulnerability review |
|
|
49
50
|
|
|
50
51
|
## Selection Rules
|
|
51
52
|
|
|
52
53
|
1. Select **exactly one** skill or agent — the most specific match wins
|
|
53
|
-
2. **Review requests**: Default to `sd-code-reviewer` agent. Only use `sd-review` skill when the
|
|
54
|
+
2. **Review requests**: Default to `sd-code-reviewer` agent. Only use `sd-review` skill when the
|
|
55
|
+
user explicitly asks for a full/comprehensive/deep review of an entire package
|
|
54
56
|
3. If nothing matches, use **default LLM behavior** and handle the request directly
|
|
55
57
|
4. Pass ARGUMENTS through as the skill/agent's input
|
|
56
58
|
|