Collections & Connectors

What's a Collection?

A collection is a named group of indexed documents from a single source. When you run indexed index create, you're creating a collection. Each collection has:

• A name (e.g., eng-tickets, team-wiki, project-docs)
• A type (files, jira, or confluence)
• A source (the folder path, Jira instance URL, or Confluence URL)
• A FAISS index containing the embedded vectors for all document chunks

Collections are independent — you can create, update, search, and delete them individually. When you search without specifying a collection, Indexed searches across all of them.

What's Stored

Each collection lives in ~/.indexed/data/collections/<name>/ and contains:

Connectors

A connector is what fetches and parses documents from a source. Indexed has three connectors:

Files

Indexes documents from a local directory. Recursively scans for all supported file types and parses them using Unstructured.

indexed index create files -c my-docs -p ./docs

Best for: project documentation, local READMEs, exported documents, any files on disk.

See the full guide: Index Your Local Docs

Jira

Fetches issues from a Jira instance using JQL (Jira Query Language) and indexes their content — summary, description, and comments.

indexed index create jira -c tickets -u https://company.atlassian.net -q "project = ENG"

Best for: ticket history, bug reports, decision records, sprint work.

See the full guide: Index Jira Tickets

Confluence

Fetches pages from a Confluence instance using CQL (Confluence Query Language) and indexes their content, including comments.

indexed index create confluence -c wiki -u https://company.atlassian.net/wiki -q "space = ENG"

Best for: team wikis, architecture docs, onboarding guides, runbooks.

See the full guide: Index Confluence Pages

Managing Collections

See the CLI Reference for all options.

What's Next