Skill Router
Analyze the request and recommend the narrowest matching skill exposed by the
current host. Do not assume Claude Code, Codex, Gemini, or DeepSeek Harness
capabilities that are not visible in the current session.
When This Skill Activates
This skill activates when you:
- Ask "which skill should I use?" or "what skill can help with...?"
- Say "use a skill" without specifying which one
- Express a need but aren't sure which skill fits
- Mention "skill router" or "help me find a skill"
Available Skills Catalog
Meta & Workflow
| Skill |
Best For |
skill-router |
Finding and selecting the right skill for a task |
create-pr |
Creating pull requests with bilingual documentation checks |
session-logger |
Saving session summaries and activity logs |
workflow-orchestrator |
Coordinating multi-skill workflows and triggers |
self-improving-agent |
Learning from completed skill workflows |
auto-trigger |
Declaring trigger relationships between skills |
Core Development
| Skill |
Best For |
commit-helper |
Writing Git commit messages, formatting commits |
code-reviewer |
Reviewing PRs, code changes, quality checks |
debugger |
Diagnosing bugs, errors, unexpected behavior |
refactoring-specialist |
Improving code structure, reducing technical debt |
Documentation & Testing
| Skill |
Best For |
documentation-engineer |
Writing README, technical docs, code documentation |
api-documenter |
Creating OpenAPI/Swagger specifications |
test-automator |
Writing tests, setting up test frameworks |
qa-expert |
Test strategy, quality gates, QA processes |
Architecture & DevOps
| Skill |
Best For |
api-designer |
Designing REST/GraphQL APIs, API architecture |
security-auditor |
Security audits, vulnerability reviews, OWASP Top 10 |
performance-engineer |
Performance optimization, speed analysis |
deployment-engineer |
CI/CD pipelines, deployment automation |
Planning & Analysis
| Skill |
Best For |
prd-planner |
Creating PRDs with persistent file-based planning |
prd-implementation-precheck |
Prechecking PRDs/specs before implementation |
architecting-solutions |
Non-PRD solution design, architecture, and requirements analysis |
planning-with-files |
Multi-step task planning, persistent file-based organization |
long-task-coordinator |
Multi-session, delegated, or resumable work that needs explicit state and recovery |
Design & UX
| Skill |
Best For |
figma-designer |
Analyzing Figma designs and producing implementation-ready visual specs/PRDs |
Routing Process
Step 1: Intent Analysis
Analyze the user's request to identify:
- Task Type: What does the user want to accomplish?
- Context: What is the working domain (web, mobile, data, etc.)?
- Complexity: Is this a simple task or complex workflow?
Step 2: Skill Matching
Match the identified intent to the most relevant skill(s) using:
- Keyword matching: Compare request keywords with skill descriptions
- Semantic similarity: Understand the meaning behind the request
- Context awareness: Consider project state and previous actions
Step 3: Interactive Clarification
If the request is ambiguous, guide the user with targeted questions:
- What is the primary goal?
- What type of output is expected?
- Are there specific constraints or preferences?
Step 4: Recommendation & Execution
Present the recommended skill with:
- Skill name and brief description
- Why it fits the current request
- Option to proceed or ask for alternatives
Routing Examples
Example 1: Clear Intent
User: "I need to review this pull request"
Router Analysis:
- Keywords: "review", "pull request"
- Intent: Code review
- Recommendation:
code-reviewer
Example 2: Ambiguous Intent
User: "Use a skill to help with my project"
Router Questions:
- What type of task are you working on?
- Are you designing, coding, testing, or documenting?
Based on answers → Recommend appropriate skill
Example 3: Multi-Skill Scenario
User: "I'm building a new API and need help with the full workflow"
Router Recommendation:
Consider using multiple skills in sequence:
api-designer - Design the API structureapi-documenter - Document endpoints with OpenAPItest-automator - Set up API testscode-reviewer - Review implementation
Interactive Question Templates
When user intent is unclear, use these question patterns:
Goal Clarification
- "What are you trying to accomplish with this task?"
- "What would the ideal outcome look like?"
Domain Identification
- "What area does this relate to: development, testing, documentation, or deployment?"
- "Are you working on code, APIs, infrastructure, or something else?"
Stage Assessment
- "What stage are you at: planning, implementing, testing, or maintaining?"
Preference Confirmation
- "Do you want a quick solution or a comprehensive approach?"
- "Are there specific tools or frameworks you're using?"
Best Practices
1. Start Broad, Then Narrow
- Begin with general category questions
- Drill down into specifics based on responses
2. Explain Your Reasoning
- Tell the user why a particular skill is recommended
- Build trust through transparency
3. Offer Alternatives
- Present the top recommendation
- Mention 1-2 alternatives if applicable
4. Handle Edge Cases
- If no skill fits perfectly, suggest the closest match
- Offer to help without a specific skill if better
5. Learn from Context
- Consider previous interactions
- Remember user preferences for future routing
Advanced Routing Patterns
Semantic Routing
Use semantic similarity when keywords don't match directly:
- "clean up my code" →
refactoring-specialist
- "make my app faster" →
performance-engineer
- "check for security issues" →
security-auditor
- "resume this interrupted workflow" →
long-task-coordinator
Multi-Skill Orchestrations
Suggest skill combinations for complex workflows:
- New Feature:
architecting-solutions → debugger → code-reviewer
- API Project:
api-designer → api-documenter → test-automator
- Production Readiness:
security-auditor → performance-engineer → deployment-engineer
Confidence Levels
Indicate confidence in recommendations:
- High: Direct keyword match, clear intent
- Medium: Semantic similarity, reasonable inference
- Low: Ambiguous request, clarification needed
Error Recovery
If the recommended skill doesn't fit:
- Acknowledge the mismatch
- Ask follow-up questions to refine understanding
- Provide alternative recommendations
- Fall back to general assistance if needed
Output Format
When recommending a skill, use this format:
## Recommended Skill: {skill-name}
{brief description of why this skill fits}
**What it does:** {one-sentence skill description}
**Best for:** {specific use cases}
---
Would you like me to activate this skill, or would you prefer to see other options?
References
1---
2name: skill-router
3description: Routes an agent request to the narrowest matching portable skill when the user asks which skill to use or needs help selecting one across supported hosts.
4allowed-tools: Read, AskUserQuestion, WebSearch, Grep
5metadata:
6 hooks:
7 after_complete:
8 - trigger: session-logger
9 mode: auto
10 reason: "Log skill routing decisions"
11---
12
13# Skill Router
14
15Analyze the request and recommend the narrowest matching skill exposed by the
16current host. Do not assume Claude Code, Codex, Gemini, or DeepSeek Harness
17capabilities that are not visible in the current session.
18
19## When This Skill Activates
20
21This skill activates when you:
22- Ask "which skill should I use?" or "what skill can help with...?"
23- Say "use a skill" without specifying which one
24- Express a need but aren't sure which skill fits
25- Mention "skill router" or "help me find a skill"
26
27## Available Skills Catalog
28
29### Meta & Workflow
30
31| Skill | Best For |
32|-------|----------|
33| `skill-router` | Finding and selecting the right skill for a task |
34| `create-pr` | Creating pull requests with bilingual documentation checks |
35| `session-logger` | Saving session summaries and activity logs |
36| `workflow-orchestrator` | Coordinating multi-skill workflows and triggers |
37| `self-improving-agent` | Learning from completed skill workflows |
38| `auto-trigger` | Declaring trigger relationships between skills |
39
40### Core Development
41
42| Skill | Best For |
43|-------|----------|
44| `commit-helper` | Writing Git commit messages, formatting commits |
45| `code-reviewer` | Reviewing PRs, code changes, quality checks |
46| `debugger` | Diagnosing bugs, errors, unexpected behavior |
47| `refactoring-specialist` | Improving code structure, reducing technical debt |
48
49### Documentation & Testing
50
51| Skill | Best For |
52|-------|----------|
53| `documentation-engineer` | Writing README, technical docs, code documentation |
54| `api-documenter` | Creating OpenAPI/Swagger specifications |
55| `test-automator` | Writing tests, setting up test frameworks |
56| `qa-expert` | Test strategy, quality gates, QA processes |
57
58### Architecture & DevOps
59
60| Skill | Best For |
61|-------|----------|
62| `api-designer` | Designing REST/GraphQL APIs, API architecture |
63| `security-auditor` | Security audits, vulnerability reviews, OWASP Top 10 |
64| `performance-engineer` | Performance optimization, speed analysis |
65| `deployment-engineer` | CI/CD pipelines, deployment automation |
66
67### Planning & Analysis
68
69| Skill | Best For |
70|-------|----------|
71| `prd-planner` | Creating PRDs with persistent file-based planning |
72| `prd-implementation-precheck` | Prechecking PRDs/specs before implementation |
73| `architecting-solutions` | Non-PRD solution design, architecture, and requirements analysis |
74| `planning-with-files` | Multi-step task planning, persistent file-based organization |
75| `long-task-coordinator` | Multi-session, delegated, or resumable work that needs explicit state and recovery |
76
77### Design & UX
78
79| Skill | Best For |
80|-------|----------|
81| `figma-designer` | Analyzing Figma designs and producing implementation-ready visual specs/PRDs |
82
83## Routing Process
84
85### Step 1: Intent Analysis
86
87Analyze the user's request to identify:
88- **Task Type**: What does the user want to accomplish?
89- **Context**: What is the working domain (web, mobile, data, etc.)?
90- **Complexity**: Is this a simple task or complex workflow?
91
92### Step 2: Skill Matching
93
94Match the identified intent to the most relevant skill(s) using:
95- **Keyword matching**: Compare request keywords with skill descriptions
96- **Semantic similarity**: Understand the meaning behind the request
97- **Context awareness**: Consider project state and previous actions
98
99### Step 3: Interactive Clarification
100
101If the request is ambiguous, guide the user with targeted questions:
102- What is the primary goal?
103- What type of output is expected?
104- Are there specific constraints or preferences?
105
106### Step 4: Recommendation & Execution
107
108Present the recommended skill with:
109- Skill name and brief description
110- Why it fits the current request
111- Option to proceed or ask for alternatives
112
113## Routing Examples
114
115### Example 1: Clear Intent
116
117**User:** "I need to review this pull request"
118
119**Router Analysis:**
120- Keywords: "review", "pull request"
121- Intent: Code review
122- **Recommendation:** `code-reviewer`
123
124### Example 2: Ambiguous Intent
125
126**User:** "Use a skill to help with my project"
127
128**Router Questions:**
1291. What type of task are you working on?
1302. Are you designing, coding, testing, or documenting?
131
132Based on answers → Recommend appropriate skill
133
134### Example 3: Multi-Skill Scenario
135
136**User:** "I'm building a new API and need help with the full workflow"
137
138**Router Recommendation:**
139Consider using multiple skills in sequence:
1401. `api-designer` - Design the API structure
1412. `api-documenter` - Document endpoints with OpenAPI
1423. `test-automator` - Set up API tests
1434. `code-reviewer` - Review implementation
144
145## Interactive Question Templates
146
147When user intent is unclear, use these question patterns:
148
149### Goal Clarification
150- "What are you trying to accomplish with this task?"
151- "What would the ideal outcome look like?"
152
153### Domain Identification
154- "What area does this relate to: development, testing, documentation, or deployment?"
155- "Are you working on code, APIs, infrastructure, or something else?"
156
157### Stage Assessment
158- "What stage are you at: planning, implementing, testing, or maintaining?"
159
160### Preference Confirmation
161- "Do you want a quick solution or a comprehensive approach?"
162- "Are there specific tools or frameworks you're using?"
163
164## Best Practices
165
166### 1. Start Broad, Then Narrow
167- Begin with general category questions
168- Drill down into specifics based on responses
169
170### 2. Explain Your Reasoning
171- Tell the user why a particular skill is recommended
172- Build trust through transparency
173
174### 3. Offer Alternatives
175- Present the top recommendation
176- Mention 1-2 alternatives if applicable
177
178### 4. Handle Edge Cases
179- If no skill fits perfectly, suggest the closest match
180- Offer to help without a specific skill if better
181
182### 5. Learn from Context
183- Consider previous interactions
184- Remember user preferences for future routing
185
186## Advanced Routing Patterns
187
188### Semantic Routing
189Use semantic similarity when keywords don't match directly:
190- "clean up my code" → `refactoring-specialist`
191- "make my app faster" → `performance-engineer`
192- "check for security issues" → `security-auditor`
193- "resume this interrupted workflow" → `long-task-coordinator`
194
195### Multi-Skill Orchestrations
196Suggest skill combinations for complex workflows:
197- **New Feature**: `architecting-solutions` → `debugger` → `code-reviewer`
198- **API Project**: `api-designer` → `api-documenter` → `test-automator`
199- **Production Readiness**: `security-auditor` → `performance-engineer` → `deployment-engineer`
200
201### Confidence Levels
202Indicate confidence in recommendations:
203- **High**: Direct keyword match, clear intent
204- **Medium**: Semantic similarity, reasonable inference
205- **Low**: Ambiguous request, clarification needed
206
207## Error Recovery
208
209If the recommended skill doesn't fit:
2101. Acknowledge the mismatch
2112. Ask follow-up questions to refine understanding
2123. Provide alternative recommendations
2134. Fall back to general assistance if needed
214
215## Output Format
216
217When recommending a skill, use this format:
218
219```markdown
220## Recommended Skill: {skill-name}
221
222{brief description of why this skill fits}
223
224**What it does:** {one-sentence skill description}
225
226**Best for:** {specific use cases}
227
228---
229
230Would you like me to activate this skill, or would you prefer to see other options?
231```
232
233## References
234
235- [AI Agent Routing: Tutorial & Best Practices](https://www.patronus.ai/ai-agent-development/ai-agent-routing)
236- [Intent Recognition and Auto-Routing in Multi-Agent Systems](https://gist.github.com/mkbctrl/a35764e99fe0c8e8c00b2358f55cd7fa)
237- [Multi-LLM Routing Strategies (AWS)](https://aws.amazon.com/blogs/machine-learning/multi-llm-routing-strategies-for-generative-ai-applications-on-aws/)
zhaono1