intelligent-routing

Name: intelligent-routing
Author: One-Man-Company

77stars16forks

How to install intelligent-routing?

one-man-company/skills-contextmanager/intelligent-routing

$npx -y skills add One-Man-Company/Skills-ContextManager --skill intelligent-routing

Installs into the current project.

Summary

Automatic agent selection and intelligent task routing. Analyzes user requests and automatically selects the best specialist agent(s) without requiring explicit user mentions.

Files · 1

View on GitHub

SKILL.md

1	# Intelligent Agent Routing
2
3	Purpose: Automatically analyze user requests and route them to the most appropriate specialist agent(s) without requiring explicit user mentions.
4
5	## Core Principle
6
7	> The AI should act as an intelligent Project Manager, analyzing each request and automatically selecting the best specialist(s) for the job.
8
9	## How It Works
10
11	### 1. Request Analysis
12
13	Before responding to ANY user request, perform automatic analysis:
14
15	```mermaid
16	graph TD
17	A[User Request: Add login] --> B[ANALYZE]
18	B --> C[Keywords]
19	B --> D[Domains]
20	B --> E[Complexity]
21	C --> F[SELECT AGENT]
22	D --> F
23	E --> F
24	F --> G[security-auditor + backend-specialist]
25	G --> H[AUTO-INVOKE with context]
26	```
27
28	### 2. Agent Selection Matrix
29
30	Use this matrix to automatically select agents:
31
32	\| User Intent \| Keywords \| Selected Agent(s) \| Auto-invoke? \|
33	\| ------------------- \| ------------------------------------------ \| ------------------------------------------- \| ------------ \|
34	\| Authentication \| "login", "auth", "signup", "password" \| `security-auditor` + `backend-specialist` \| ✅ YES \|
35	\| UI Component \| "button", "card", "layout", "style" \| `frontend-specialist` \| ✅ YES \|
36	\| Mobile UI \| "screen", "navigation", "touch", "gesture" \| `mobile-developer` \| ✅ YES \|
37	\| API Endpoint \| "endpoint", "route", "API", "POST", "GET" \| `backend-specialist` \| ✅ YES \|
38	\| Database \| "schema", "migration", "query", "table" \| `database-architect` + `backend-specialist` \| ✅ YES \|
39	\| Bug Fix \| "error", "bug", "not working", "broken" \| `debugger` \| ✅ YES \|
40	\| Test \| "test", "coverage", "unit", "e2e" \| `test-engineer` \| ✅ YES \|
41	\| Deployment \| "deploy", "production", "CI/CD", "docker" \| `devops-engineer` \| ✅ YES \|
42	\| Security Review \| "security", "vulnerability", "exploit" \| `security-auditor` + `penetration-tester` \| ✅ YES \|
43	\| Performance \| "slow", "optimize", "performance", "speed" \| `performance-optimizer` \| ✅ YES \|
44	\| Product Def \| "requirements", "user story", "backlog", "MVP" \| `product-owner` \| ✅ YES \|
45	\| New Feature \| "build", "create", "implement", "new app" \| `orchestrator` → multi-agent \| ⚠️ ASK FIRST \|
46	\| Complex Task \| Multiple domains detected \| `orchestrator` → multi-agent \| ⚠️ ASK FIRST \|
47
48	### 3. Automatic Routing Protocol
49
50	## TIER 0 - Automatic Analysis (ALWAYS ACTIVE)
51
52	Before responding to ANY request:
53
54	```javascript
55	// Pseudo-code for decision tree
56	function analyzeRequest(userMessage) {
57	// 1. Classify request type
58	const requestType = classifyRequest(userMessage);
59
60	// 2. Detect domains
61	const domains = detectDomains(userMessage);
62
63	// 3. Determine complexity
64	const complexity = assessComplexity(domains);
65
66	// 4. Select agent(s)
67	if (complexity === "SIMPLE" && domains.length === 1) {
68	return selectSingleAgent(domains[0]);
69	} else if (complexity === "MODERATE" && domains.length <= 2) {
70	return selectMultipleAgents(domains);
71	} else {
72	return "orchestrator"; // Complex task
73	}
74	}
75	```
76
77	## 4. Response Format
78
79	When auto-selecting an agent, inform the user concisely:
80
81	```markdown
82	🤖 Applying knowledge of `@security-auditor` + `@backend-specialist`...
83
84	[Proceed with specialized response]
85	```
86
87	Benefits:
88
89	- ✅ User sees which expertise is being applied
90	- ✅ Transparent decision-making
91	- ✅ Still automatic (no /commands needed)
92
93	## Domain Detection Rules
94
95	### Single-Domain Tasks (Auto-invoke Single Agent)
96
97	\| Domain \| Patterns \| Agent \|
98	\| --------------- \| ------------------------------------------ \| ----------------------- \|
99	\| Security \| auth, login, jwt, password, hash, token \| `security-auditor` \|
100	\| Frontend \| component, react, vue, css, html, tailwind \| `frontend-specialist` \|
101	\| Backend \| api, server, express, fastapi, node \| `backend-specialist` \|
102	\| Mobile \| react native, flutter, ios, android, expo \| `mobile-developer` \|
103	\| Database \| prisma, sql, mongodb, schema, migration \| `database-architect` \|
104	\| Testing \| test, jest, vitest, playwright, cypress \| `test-engineer` \|
105	\| DevOps \| docker, kubernetes, ci/cd, pm2, nginx \| `devops-engi

People who liked this also viewed

6 picks

Type

intelligent-routing

77stars16forks

View on GitHub

How to install intelligent-routing?

one-man-company/skills-contextmanager/intelligent-routing

$npx -y skills add One-Man-Company/Skills-ContextManager --skill intelligent-routing

Installs into the current project.

Summary

Automatic agent selection and intelligent task routing. Analyzes user requests and automatically selects the best specialist agent(s) without requiring explicit user mentions.

Files · 1

View on GitHub

SKILL.md

1	# Intelligent Agent Routing
2
3	Purpose: Automatically analyze user requests and route them to the most appropriate specialist agent(s) without requiring explicit user mentions.
4
5	## Core Principle
6
7	> The AI should act as an intelligent Project Manager, analyzing each request and automatically selecting the best specialist(s) for the job.
8
9	## How It Works
10
11	### 1. Request Analysis
12
13	Before responding to ANY user request, perform automatic analysis:
14
15	```mermaid
16	graph TD
17	A[User Request: Add login] --> B[ANALYZE]
18	B --> C[Keywords]
19	B --> D[Domains]
20	B --> E[Complexity]
21	C --> F[SELECT AGENT]
22	D --> F
23	E --> F
24	F --> G[security-auditor + backend-specialist]
25	G --> H[AUTO-INVOKE with context]
26	```
27
28	### 2. Agent Selection Matrix
29
30	Use this matrix to automatically select agents:
31
32	\| User Intent \| Keywords \| Selected Agent(s) \| Auto-invoke? \|
33	\| ------------------- \| ------------------------------------------ \| ------------------------------------------- \| ------------ \|
34	\| Authentication \| "login", "auth", "signup", "password" \| `security-auditor` + `backend-specialist` \| ✅ YES \|
35	\| UI Component \| "button", "card", "layout", "style" \| `frontend-specialist` \| ✅ YES \|
36	\| Mobile UI \| "screen", "navigation", "touch", "gesture" \| `mobile-developer` \| ✅ YES \|
37	\| API Endpoint \| "endpoint", "route", "API", "POST", "GET" \| `backend-specialist` \| ✅ YES \|
38	\| Database \| "schema", "migration", "query", "table" \| `database-architect` + `backend-specialist` \| ✅ YES \|
39	\| Bug Fix \| "error", "bug", "not working", "broken" \| `debugger` \| ✅ YES \|
40	\| Test \| "test", "coverage", "unit", "e2e" \| `test-engineer` \| ✅ YES \|
41	\| Deployment \| "deploy", "production", "CI/CD", "docker" \| `devops-engineer` \| ✅ YES \|
42	\| Security Review \| "security", "vulnerability", "exploit" \| `security-auditor` + `penetration-tester` \| ✅ YES \|
43	\| Performance \| "slow", "optimize", "performance", "speed" \| `performance-optimizer` \| ✅ YES \|
44	\| Product Def \| "requirements", "user story", "backlog", "MVP" \| `product-owner` \| ✅ YES \|
45	\| New Feature \| "build", "create", "implement", "new app" \| `orchestrator` → multi-agent \| ⚠️ ASK FIRST \|
46	\| Complex Task \| Multiple domains detected \| `orchestrator` → multi-agent \| ⚠️ ASK FIRST \|
47
48	### 3. Automatic Routing Protocol
49
50	## TIER 0 - Automatic Analysis (ALWAYS ACTIVE)
51
52	Before responding to ANY request:
53
54	```javascript
55	// Pseudo-code for decision tree
56	function analyzeRequest(userMessage) {
57	// 1. Classify request type
58	const requestType = classifyRequest(userMessage);
59
60	// 2. Detect domains
61	const domains = detectDomains(userMessage);
62
63	// 3. Determine complexity
64	const complexity = assessComplexity(domains);
65
66	// 4. Select agent(s)
67	if (complexity === "SIMPLE" && domains.length === 1) {
68	return selectSingleAgent(domains[0]);
69	} else if (complexity === "MODERATE" && domains.length <= 2) {
70	return selectMultipleAgents(domains);
71	} else {
72	return "orchestrator"; // Complex task
73	}
74	}
75	```
76
77	## 4. Response Format
78
79	When auto-selecting an agent, inform the user concisely:
80
81	```markdown
82	🤖 Applying knowledge of `@security-auditor` + `@backend-specialist`...
83
84	[Proceed with specialized response]
85	```
86
87	Benefits:
88
89	- ✅ User sees which expertise is being applied
90	- ✅ Transparent decision-making
91	- ✅ Still automatic (no /commands needed)
92
93	## Domain Detection Rules
94
95	### Single-Domain Tasks (Auto-invoke Single Agent)
96
97	\| Domain \| Patterns \| Agent \|
98	\| --------------- \| ------------------------------------------ \| ----------------------- \|
99	\| Security \| auth, login, jwt, password, hash, token \| `security-auditor` \|
100	\| Frontend \| component, react, vue, css, html, tailwind \| `frontend-specialist` \|
101	\| Backend \| api, server, express, fastapi, node \| `backend-specialist` \|
102	\| Mobile \| react native, flutter, ios, android, expo \| `mobile-developer` \|
103	\| Database \| prisma, sql, mongodb, schema, migration \| `database-architect` \|
104	\| Testing \| test, jest, vitest, playwright, cypress \| `test-engineer` \|
105	\| DevOps \| docker, kubernetes, ci/cd, pm2, nginx \| `devops-engi

People who liked this also viewed

6 picks

Type