Existing Collections

Create a new indexed collection for semantic search.

Existing Collections

!indexed index inspect 2>/dev/null || echo "No collections yet."

Available Source Types

• files — Local files and directories (most common for codebases)
• jira — Jira issues (requires credentials in .env)
• confluence — Confluence pages (requires credentials in .env)

Commands

From local files

indexed index create files -c <collection-name> -p <path>

Common options:

• -c, --collection — Collection name (required)
• -p, --path — Root directory or file path
• --include <regex> — Regex patterns to include (repeatable)
• --exclude <regex> — Regex patterns to exclude (repeatable)
• --force — Overwrite existing collection with same name
• --use-cache/--no-cache — Enable/disable caching (default: enabled)
• --fail-fast/--no-fail-fast — Stop on first read error

Example for a Python codebase:

indexed index create files -c my-code -p ./src \
  --include '\.py$' \
  --exclude '__pycache__|\.pyc$'

From Jira

indexed index create jira -c <collection-name>

From Confluence

indexed index create confluence -c <collection-name>

After Creation

Verify the collection was created:

indexed index inspect <collection-name>

Then use /index-use or mcp__indexed__search to query it.