CLI Reference

Complete reference for all indexed CLI commands.

`indexed index create`

Create a new collection from a data source.

`indexed index create files`

indexed index create files -c <name> -p <path> [flags]

Flag	Type	Required	Default	Description
`-c, --collection`	string	Yes	—	Collection name
`-p, --path`	string	Yes	—	Path to directory to index
`--include`	string	No	—	Regex pattern for files to include (repeatable)
`--exclude`	string	No	—	Regex pattern for files to exclude (repeatable)
`--force`	bool	No	`false`	Overwrite existing collection
`--fail-fast`	bool	No	`false`	Stop on first parse error
`--no-fail-fast`	bool	No	`true`	Skip unparseable files and continue
`--use-cache`	bool	No	`true`	Reuse cached parsed documents
`--no-cache`	bool	No	`false`	Re-parse all documents

Example:

indexed index create files -c my-docs -p ./docs --include ".*\.md$" --force

`indexed index create jira`

indexed index create jira -c <name> -u <url> -q <jql> [flags]

Flag	Type	Required	Default	Description
`-c, --collection`	string	Yes	—	Collection name
`-u, --url`	string	Yes	—	Jira instance URL
`-q, --jql, --query`	string	Yes	—	JQL query to filter issues
`--force`	bool	No	`false`	Overwrite existing collection
`--use-cache`	bool	No	`true`	Reuse cached data
`--no-cache`	bool	No	`false`	Re-fetch all data

Environment variables: JIRA_API_TOKEN, JIRA_EMAIL (Cloud only)

Example:

indexed index create jira -c eng-tickets \
  -u https://company.atlassian.net \
  -q "project = ENG AND created >= -90d"

`indexed index create confluence`

indexed index create confluence -c <name> -u <url> -q <cql> [flags]

Flag	Type	Required	Default	Description
`-c, --collection`	string	Yes	—	Collection name
`-u, --url`	string	Yes	—	Confluence instance URL (include `/wiki` for Cloud)
`-q, --cql, --query`	string	Yes	—	CQL query to filter pages
`--read-all-comments`	bool	No	`true`	Index all comment levels
`--first-level-comments`	bool	No	`false`	Index only top-level comments
`--force`	bool	No	`false`	Overwrite existing collection
`--use-cache`	bool	No	`true`	Reuse cached data
`--no-cache`	bool	No	`false`	Re-fetch all data

Environment variables: CONFLUENCE_API_TOKEN, CONFLUENCE_EMAIL (Cloud only)

Example:

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

`indexed index search`

Search across collections.

indexed index search "<query>" [flags]

Flag	Type	Required	Default	Description
`-c, --collection`	string	No	all	Search a specific collection
`-l, --limit`	int	No	`5`	Maximum number of results
`--compact`	bool	No	`false`	Compact output format
`--no-content`	bool	No	`false`	Show metadata only, no text

Examples:

# Search all collections
indexed index search "authentication methods"

# Search one collection, limit results
indexed index search "rate limiting" -c eng-tickets -l 10

# Compact output
indexed index search "deploy" --compact

`indexed index inspect`

View collection metadata and statistics.

indexed index inspect [name] [flags]

Flag	Type	Required	Default	Description
`name`	string	No	all	Specific collection to inspect
`--verbose`	bool	No	`false`	Show extended details
`--json`	bool	No	`false`	Output as JSON

Examples:

# All collections
indexed index inspect

# Single collection as JSON
indexed index inspect my-docs --json

`indexed index update`

Refresh collections with latest data from their sources.

indexed index update [name]

Flag	Type	Required	Default	Description
`name`	string	No	all	Specific collection to update (omit for all)

Examples:

# Update one collection
indexed index update eng-tickets

# Update all collections
indexed index update

`indexed index delete`

Permanently remove a collection.

indexed index delete <name> [flags]

Flag	Type	Required	Default	Description
`name`	string	Yes	—	Collection to delete
`--force`	bool	No	`false`	Skip confirmation prompt

Deletion is permanent. The FAISS index and all associated data are removed.

`indexed config init`

Initialize the Indexed workspace and configuration.

indexed config init [flags]

Flag	Type	Required	Default	Description
`--yes`	bool	No	`false`	Accept all defaults, no prompts

`indexed config inspect`

View current configuration.

indexed config inspect [flags]

Flag	Type	Required	Default	Description
`--json`	bool	No	`false`	Output as JSON

`indexed config set`

Set a configuration value.

indexed config set <key> <value> [flags]

Flag	Type	Required	Default	Description
`key`	string	Yes	—	Config key path (e.g., `core.v1.indexing.chunk_size`)
`value`	string	Yes	—	New value
`--dry-run`	bool	No	`false`	Preview the change without applying it

`indexed config delete`

Remove a configuration value (reset to default).

indexed config delete <key>

`indexed config validate`

Validate the current configuration file.

indexed config validate

`indexed mcp run`

Start the MCP server.

indexed mcp run [flags]

Flag	Type	Required	Default	Description
`--transport`	string	No	`stdio`	Transport type: `stdio`, `http`, `sse`, `streamable-http`
`--port`	int	No	`8000`	Port for HTTP/SSE transports
`--host`	string	No	`127.0.0.1`	Host to bind for HTTP/SSE transports

Examples:

# Default stdio (for Claude Desktop, Cursor)
indexed mcp run

# HTTP server
indexed mcp run --transport http --port 8000

# HTTP on all interfaces (Docker)
indexed mcp run --transport http --host 0.0.0.0 --port 8000

`indexed mcp dev`

Start the MCP server with the MCP Inspector for development and debugging.

indexed mcp dev

CLI Reference

On this page