How do I install the RAG Index skill?

Run npx skillmds@latest add etalab-ia/rag-index-2 in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is RAG Index safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with RAG Index?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is RAG Index free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

Who published RAG Index?

etalab-ia (@etalab-ia) published this skill. Their other Agent Skills are listed on their SkillMD profile.

RAG Index by etalab-ia — Agent Skill

Name: RAG Index
Author: etalab-ia

RAG Index

Index a document corpus for semantic search. Use when the user wants to set up a knowledge base, create a searchable index from markdown documents, or enable semantic search. Triggers on keywords like "index documents", "create knowledge base", "setup search", "semantic search".

etalab-ia Updated Sep 22, 2026

File contents

rag-index Skill

Index a document corpus for semantic search. This skill uses qmd under the hood.

Prerequisites

Install qmd:

bun install -g @tobilu/qmd

Verify installation:

qmd --version

Documents must be in markdown format. Use /rag-parse first to convert PDFs/DOCX.

Response Requirements

When using this skill, always include:

Collection name that will be created/used (e.g. my-project)
Source documents path being indexed (e.g. ./docs)
Where the index/collection is stored (qmd collection location), and a verification command (qmd status)

Never omit storage location details in your final answer.

Workflow

1. Verify qmd

qmd --version

If not installed, see Prerequisites above.

2. Create Collection

qmd collection add <path> --name <name>

Arguments:

Argument	Type	Default	Description
`path`	string	required	Path to document directory
`name`	string	derived	Collection name (defaults to directory name)

Examples:

qmd collection add ./docs --name my-project
qmd collection add ~/Documents/api-docs --name api-docs

3. Add Context (Optional)

Add semantic context to help with retrieval:

qmd context add qmd://<name> "<description>"

Examples:

qmd context add qmd://api-docs "API documentation for the project"
qmd context add qmd://contracts "Legal contracts and agreements"

4. Generate Embeddings

qmd embed

This creates vector embeddings for all documents in the collection.

5. Verify Setup

qmd status

Should show your collection with document count.

Full Example

# Create collection
qmd collection add ./docs --name my-project

# Add context
qmd context add qmd://my-project "Project documentation and guides"

# Generate embeddings
qmd embed

# Verify
qmd status

Arguments

Argument	Type	Default	Description
`path`	string	required	Path to document directory
`name`	string	derived	Collection name
`context`	string	asked	Context description for retrieval

Troubleshooting

"sqlite-vec is not available"

qmd requires SQLite with extension loading support. Solutions:

macOS: brew install sqlite
Linux: Build sqlite-vec from source
Docker: Use official qmd Docker image

See docs/qmd-setup.md for detailed instructions.

Other Issues

If embeddings fail:

Check qmd status: qmd status
Verify documents exist: ls -la <path>
Check for empty files: find <path> -empty

Next Steps

After setting up the knowledge base, use /rag-search to search your documents.

Provider-Specific Notes

qmd (current)

Storage: Local SQLite with sqlite-vec extension
Embeddings: Local model (no API key required)
Best for: Small to medium corpora, offline usage

pinecone (planned)

Storage: Pinecone cloud
Embeddings: OpenAI or custom embeddings
Best for: Large-scale production deployments
Setup: Requires PINECONE_API_KEY environment variable

weaviate (planned)

Storage: Weaviate instance (self-hosted or cloud)
Embeddings: Configurable
Best for: Enterprise deployments with hybrid search
Setup: Requires WEAVIATE_URL and optionally WEAVIATE_API_KEY

---name: rag-index-2description: Index a document corpus for semantic search. Use when the user wants to set up a knowledge base, create a searchable index from markdown documents, or enable semantic search. Triggers on keywords like "index documents", "create knowledge base", "setup search", "semantic search".license: MIT---# rag-index SkillIndex a document corpus for semantic search. This skill uses **qmd** under the hood.## PrerequisitesInstall qmd:```bashbun install -g @tobilu/qmd```Verify installation:```bashqmd --version```Documents must be in markdown format. Use `/rag-parse` first to convert PDFs/DOCX.## Response RequirementsWhen using this skill, always include:1. **Collection name** that will be created/used (e.g. `my-project`)2. **Source documents path** being indexed (e.g. `./docs`)3. **Where the index/collection is stored** (qmd collection location), and a verification command (`qmd status`)Never omit storage location details in your final answer.## Workflow### 1. Verify qmd```bashqmd --version```If not installed, see Prerequisites above.### 2. Create Collection```bashqmd collection add <path> --name <name>```Arguments:| Argument | Type | Default | Description ||----------|------|---------|-------------|| `path` | string | required | Path to document directory || `name` | string | derived | Collection name (defaults to directory name) |Examples:```bashqmd collection add ./docs --name my-projectqmd collection add ~/Documents/api-docs --name api-docs```### 3. Add Context (Optional)Add semantic context to help with retrieval:```bashqmd context add qmd://<name> "<description>"```Examples:```bashqmd context add qmd://api-docs "API documentation for the project"qmd context add qmd://contracts "Legal contracts and agreements"```### 4. Generate Embeddings```bashqmd embed```This creates vector embeddings for all documents in the collection.### 5. Verify Setup```bashqmd status```Should show your collection with document count.## Full Example```bash# Create collectionqmd collection add ./docs --name my-project# Add contextqmd context add qmd://my-project "Project documentation and guides"# Generate embeddingsqmd embed# Verifyqmd status```## Arguments| Argument | Type | Default | Description ||----------|------|---------|-------------|| `path` | string | required | Path to document directory || `name` | string | derived | Collection name || `context` | string | asked | Context description for retrieval |## Troubleshooting### "sqlite-vec is not available"qmd requires SQLite with extension loading support. Solutions:- **macOS**: `brew install sqlite`- **Linux**: Build sqlite-vec from source- **Docker**: Use official qmd Docker imageSee docs/qmd-setup.md for detailed instructions.### Other IssuesIf embeddings fail:1. Check qmd status: `qmd status`2. Verify documents exist: `ls -la <path>`3. Check for empty files: `find <path> -empty`## Next StepsAfter setting up the knowledge base, use `/rag-search` to search your documents.## Provider-Specific Notes### qmd (current)- Storage: Local SQLite with sqlite-vec extension- Embeddings: Local model (no API key required)- Best for: Small to medium corpora, offline usage### pinecone (planned)- Storage: Pinecone cloud- Embeddings: OpenAI or custom embeddings- Best for: Large-scale production deployments- Setup: Requires `PINECONE_API_KEY` environment variable### weaviate (planned)- Storage: Weaviate instance (self-hosted or cloud)- Embeddings: Configurable- Best for: Enterprise deployments with hybrid search- Setup: Requires `WEAVIATE_URL` and optionally `WEAVIATE_API_KEY`

Frequently asked questions