rag-agent-creator

Name: rag-agent-creator
Author: evan043

60stars6forks

How to install rag-agent-creator?

evan043/claude-cli-advanced-starter-pack/rag-agent-creator

$npx -y skills add evan043/claude-cli-advanced-starter-pack --skill rag-agent-creator

Installs into the current project.

Summary

Create RAG-enhanced Claude Code agents with structured knowledge bases, retrieval strategies, and domain expertise

Files · 1

View on GitHub

SKILL.md

1	# RAG Agent Creator Skill
2
3	Expert-level RAG (Retrieval-Augmented Generation) agent creation for Claude Code CLI.
4
5	## When to Use This Skill
6
7	This skill is automatically invoked when:
8
9	- Creating agents that need domain-specific expertise
10	- Setting up knowledge bases for agent reference
11	- Designing retrieval strategies for agent context
12	- Building multi-source knowledge systems
13
14	## What is RAG for Claude Code?
15
16	RAG (Retrieval-Augmented Generation) agents combine:
17
18	1. Retrieval - Find relevant context from knowledge base
19	2. Augmentation - Inject context into agent prompt
20	3. Generation - Agent produces informed responses
21
22	### When to Add RAG Capabilities
23
24	Add RAG when your agent needs:
25
26	- Domain-specific expertise - Specialized knowledge
27	- Large knowledge base - Multiple docs, patterns, examples
28	- Consistent retrieval - Predictable keyword-to-file mappings
29	- Protocol-based workflows - Documented procedures
30
31	## RAG Directory Structure
32
33	```
34	.claude/skills/{agent-name}/
35	skill.md # Main skill definition
36	context/ # Knowledge base
37	README.md # Knowledge index
38	PROTOCOL.md # Agent protocols (optional)
39	architecture/ # Architecture docs
40	patterns/ # Code/workflow patterns
41	examples/ # Example implementations
42	rag/ # RAG-specific config
43	README.md # Retrieval strategy
44	RETRIEVAL_MATRIX.md # Keyword mappings
45	workflows/
46	README.md
47	rag-orchestrator-agent.md # RAG coordination
48	rag-implementation-agent.md # RAG implementation
49	```
50
51	## Knowledge Base Design
52
53	### Key Principles
54
55	1. Atomic Files - One concept per file for precise retrieval
56	2. Clear Naming - Descriptive names (e.g., `basetab-pattern.md`)
57	3. Indexed - Maintain README.md with file descriptions
58	4. Cross-Referenced - Link related documents
59
60	### Example Knowledge Organization
61
62	```
63	context/
64	README.md # Index: lists all files
65	PROTOCOL.md # Core rules for agent
66
67	architecture/
68	README.md # Architecture overview
69	database-schema.md # Database structure
70	api-design.md # API patterns
71
72	patterns/
73	README.md # Pattern index
74	error-handling.md # Error patterns
75	validation.md # Input validation
76	caching.md # Cache strategies
77
78	examples/
79	README.md # Example index
80	feature-implementation.md # Complete feature
81	api-endpoint.md # API example
82	test-suite.md # Testing example
83	```
84
85	## Retrieval Patterns
86
87	### Pattern 1: Keyword Matching
88
89	Map keywords to specific knowledge files:
90
91	```markdown
92	## Retrieval Matrix
93
94	\| Keyword Pattern \| Files to Load \| Priority \|
95	\|-----------------\|---------------\|----------\|
96	\| "authentication" \| `patterns/auth.md`, `examples/auth.md` \| High \|
97	\| "database" \| `architecture/database.md`, `patterns/repository.md` \| Medium \|
98	\| "testing" \| `patterns/testing.md`, `examples/test-example.md` \| High \|
99	\| "error" \| `patterns/error-handling.md` \| Medium \|
100	```
101
102	### Pattern 2: Hierarchical Retrieval
103
104	Start specific, broaden if needed:
105
106	1. Load specific pattern file (e.g., `patterns/auth.md`)
107	2. If insufficient, load architecture file (e.g., `architecture/security.md`)
108	3. If still needed, load examples (e.g., `examples/auth-implementation.md`)
109
110	### Pattern 3: Category-Based
111
112	Group knowledge by domain:
113
114	```markdown
115	## Categories
116
117	### Backend
118	- `architecture/api-design.md`
119	- `patterns/repository.md`
120	- `patterns/service-layer.md`
121
122	### Frontend
123	- `architecture/component-structure.md`
124	- `patterns/state-management.md`
125	- `patterns/form-handling.md`
126
127	### Testing
128	- `patterns/unit-testing.md`
129	- `patterns/e2e-testing.md`
130	- `examples/test-suite.md`
131	```
132
133	### Pattern 4: Priority-Based
134
135	Load based on task importance:
136
137	```markdown
138	## Priority Levels
139
140	### Always Load (P1)
141	- `PROTOCOL.md` - Core rules
142	- `architecture/README.md` - Overview
143
144	### Load for Implementation (P2)
145	- Relevant `patterns/*.md`
146	- Relevant `architecture/*.md`
147
148	### Load for Examples (P3)
149	- `examples/*.md` - When needed
150	```
151
152	## Creating a RAG Agent
153
154	### Step 1: Define Knowledge Domain
155
156	What expertise does this agent need?
157
158	```markdown
159	## Domain: E-commerce Backend
160
161	### Topics Covered
162	- Product management
163	- Order processing
164	- Inventory tracking
165	- Payment integration
166	- User authentication
167	```
168
169	### Step 2: Create Knowledge Structure
170
171	```bash
172	mkdir -p .claude/skills/my-agent/context/{architecture,patterns,examples,rag}
173	```
174
175	### Step 3: Write Knowledge Files
176
177	Each file should be self-contained:
178
179	```markdown
180	# Pattern: Repository
181
182	## Purpose
183	Encapsulate data access logic.
184
185	## Implementation
186	[Code examples]
187
188	## When to Use
189	[Scenarios]
190
191	## Anti

People who liked this also viewed

6 picks

Type

lark-openapi-explorer飞书/Lark 原生 OpenAPI 探索：从官方文档库中挖掘未经 CLI 封装的原生 OpenAPI 接口。当用户的需求无法被现有 lark-* skill 或 lark-cli 已注册命令满足，需要查找并调用原生飞书 OpenAPI 时使用。SkillsJul 2026378k16k
lark-skill-maker创建 lark-cli 的自定义 Skill。当用户需要把飞书 API 操作封装成可复用的 Skill（包装原子 API 或编排多步流程）时使用。SkillsJul 2026377k16k
azure-cost-optimizationOfficial agent plugin providing skills and MCP server configurations for Azure scenarios.SkillsJul 2026209k1.3k
vercel-react-native-skillsVercel's official collection of agent skillsSkillsJul 2026172k29k
mcp-builderGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate externaSkillsJul 202693k163k
gws-sharedgws CLI: Shared patterns for authentication, global flags, and output formatting.SkillsJul 202653k30k

rag-agent-creator

60stars6forks

View on GitHub

How to install rag-agent-creator?

evan043/claude-cli-advanced-starter-pack/rag-agent-creator

$npx -y skills add evan043/claude-cli-advanced-starter-pack --skill rag-agent-creator

Installs into the current project.

Summary

Create RAG-enhanced Claude Code agents with structured knowledge bases, retrieval strategies, and domain expertise

Files · 1

View on GitHub

SKILL.md

1	# RAG Agent Creator Skill
2
3	Expert-level RAG (Retrieval-Augmented Generation) agent creation for Claude Code CLI.
4
5	## When to Use This Skill
6
7	This skill is automatically invoked when:
8
9	- Creating agents that need domain-specific expertise
10	- Setting up knowledge bases for agent reference
11	- Designing retrieval strategies for agent context
12	- Building multi-source knowledge systems
13
14	## What is RAG for Claude Code?
15
16	RAG (Retrieval-Augmented Generation) agents combine:
17
18	1. Retrieval - Find relevant context from knowledge base
19	2. Augmentation - Inject context into agent prompt
20	3. Generation - Agent produces informed responses
21
22	### When to Add RAG Capabilities
23
24	Add RAG when your agent needs:
25
26	- Domain-specific expertise - Specialized knowledge
27	- Large knowledge base - Multiple docs, patterns, examples
28	- Consistent retrieval - Predictable keyword-to-file mappings
29	- Protocol-based workflows - Documented procedures
30
31	## RAG Directory Structure
32
33	```
34	.claude/skills/{agent-name}/
35	skill.md # Main skill definition
36	context/ # Knowledge base
37	README.md # Knowledge index
38	PROTOCOL.md # Agent protocols (optional)
39	architecture/ # Architecture docs
40	patterns/ # Code/workflow patterns
41	examples/ # Example implementations
42	rag/ # RAG-specific config
43	README.md # Retrieval strategy
44	RETRIEVAL_MATRIX.md # Keyword mappings
45	workflows/
46	README.md
47	rag-orchestrator-agent.md # RAG coordination
48	rag-implementation-agent.md # RAG implementation
49	```
50
51	## Knowledge Base Design
52
53	### Key Principles
54
55	1. Atomic Files - One concept per file for precise retrieval
56	2. Clear Naming - Descriptive names (e.g., `basetab-pattern.md`)
57	3. Indexed - Maintain README.md with file descriptions
58	4. Cross-Referenced - Link related documents
59
60	### Example Knowledge Organization
61
62	```
63	context/
64	README.md # Index: lists all files
65	PROTOCOL.md # Core rules for agent
66
67	architecture/
68	README.md # Architecture overview
69	database-schema.md # Database structure
70	api-design.md # API patterns
71
72	patterns/
73	README.md # Pattern index
74	error-handling.md # Error patterns
75	validation.md # Input validation
76	caching.md # Cache strategies
77
78	examples/
79	README.md # Example index
80	feature-implementation.md # Complete feature
81	api-endpoint.md # API example
82	test-suite.md # Testing example
83	```
84
85	## Retrieval Patterns
86
87	### Pattern 1: Keyword Matching
88
89	Map keywords to specific knowledge files:
90
91	```markdown
92	## Retrieval Matrix
93
94	\| Keyword Pattern \| Files to Load \| Priority \|
95	\|-----------------\|---------------\|----------\|
96	\| "authentication" \| `patterns/auth.md`, `examples/auth.md` \| High \|
97	\| "database" \| `architecture/database.md`, `patterns/repository.md` \| Medium \|
98	\| "testing" \| `patterns/testing.md`, `examples/test-example.md` \| High \|
99	\| "error" \| `patterns/error-handling.md` \| Medium \|
100	```
101
102	### Pattern 2: Hierarchical Retrieval
103
104	Start specific, broaden if needed:
105
106	1. Load specific pattern file (e.g., `patterns/auth.md`)
107	2. If insufficient, load architecture file (e.g., `architecture/security.md`)
108	3. If still needed, load examples (e.g., `examples/auth-implementation.md`)
109
110	### Pattern 3: Category-Based
111
112	Group knowledge by domain:
113
114	```markdown
115	## Categories
116
117	### Backend
118	- `architecture/api-design.md`
119	- `patterns/repository.md`
120	- `patterns/service-layer.md`
121
122	### Frontend
123	- `architecture/component-structure.md`
124	- `patterns/state-management.md`
125	- `patterns/form-handling.md`
126
127	### Testing
128	- `patterns/unit-testing.md`
129	- `patterns/e2e-testing.md`
130	- `examples/test-suite.md`
131	```
132
133	### Pattern 4: Priority-Based
134
135	Load based on task importance:
136
137	```markdown
138	## Priority Levels
139
140	### Always Load (P1)
141	- `PROTOCOL.md` - Core rules
142	- `architecture/README.md` - Overview
143
144	### Load for Implementation (P2)
145	- Relevant `patterns/*.md`
146	- Relevant `architecture/*.md`
147
148	### Load for Examples (P3)
149	- `examples/*.md` - When needed
150	```
151
152	## Creating a RAG Agent
153
154	### Step 1: Define Knowledge Domain
155
156	What expertise does this agent need?
157
158	```markdown
159	## Domain: E-commerce Backend
160
161	### Topics Covered
162	- Product management
163	- Order processing
164	- Inventory tracking
165	- Payment integration
166	- User authentication
167	```
168
169	### Step 2: Create Knowledge Structure
170
171	```bash
172	mkdir -p .claude/skills/my-agent/context/{architecture,patterns,examples,rag}
173	```
174
175	### Step 3: Write Knowledge Files
176
177	Each file should be self-contained:
178
179	```markdown
180	# Pattern: Repository
181
182	## Purpose
183	Encapsulate data access logic.
184
185	## Implementation
186	[Code examples]
187
188	## When to Use
189	[Scenarios]
190
191	## Anti

People who liked this also viewed

6 picks

Type