Skip to main content
infino is the command-line interface to the retrieval engine: SQL, full-text (BM25), and vector search over a single copy of your data on object storage, run straight from your terminal — or driven by a coding agent. It wraps the same engine as the Python, Node, and Rust SDKs; there’s nothing extra to run.

​
Install

brew install infino-ai/tap/infino-cli
All install the infino binary.

​
Connect

Every command targets a storage location with --uri (or the INFINO_URI environment variable):
--uriStorage
memory://in-process, ephemeral
file://<path>local disk
s3://<bucket>/<prefix>Amazon S3 (or S3-compatible)
az://<container>/<prefix>Azure Blob
S3/Azure credentials are read from the ambient environment (AWS_*, AZURE_*).

​
Quickstart

# Create a table and load its first rows (body full-text indexed)
infino create-table docs --uri file://./data --schema schema.yaml --fts body --file seed.ndjson

# Append more rows
infino ingest docs --uri file://./data --file more.ndjson --format ndjson

# Search
infino bm25-search docs body "object storage" -k 10 --uri file://./data
infino query "SELECT id, body FROM docs LIMIT 10" --uri file://./data --output json
A table becomes durable on its first commit, so create-table loads initial rows too — from a Parquet file (--from-parquet, which also infers the schema) or a YAML schema plus --file.

​
Commands

CommandDescription
create-tableCreate a table and load initial rows; declare --fts / --vector indexes
ingestAppend rows from Parquet or NDJSON (file or stdin)
bm25-searchRanked keyword (BM25) search
vector-searchVector similarity (kNN) search
token-match / exact-matchUnranked token / exact-value match
queryRun SQL, including the bm25_search() / vector_search() table functions
tables / describeList tables / show a table’s schema
update / deleteChange rows matching a --where SQL predicate
optimizeCompact a table
skills installInstall the bundled agent skills for Claude Code / Cursor
Run infino <command> --help for the full flags. Every row-returning command takes --output table (default), json, or csv. The CLI does not embed text — embed your query with your own model and pass the vector as a JSON array (or - for stdin): 
infino vector-search docs embedding --vector-file query.json -k 10 --uri file://./data
Give the column a vector index when you create the table: --vector embedding:384:256:cosine (column:dim:centroids:metric).

​
Agent skills

infino skills install writes skill files into ~/.claude/skills so coding agents (Claude Code, Cursor) can drive the CLI in natural language: 
infino skills install
infino skills status
This complements the MCP server: skills are for shell-native agents, MCP is for tool-calling agents.

​
Learn more

Last modified on July 1, 2026