Quick Start¶

From nothing to a running, verified agent-utilities in a few minutes. This is the fast path; for the full config-complete walkthrough (secrets, profiles, multi-node) see the Self-Setup guide, and for the database environment see the Stardog + pg-age recipe.

TL;DR — zero-infra, 4 commands¶

pip install agent-utilities[all]
setup-config generate --profile tiny        # complete config.json (every option)
graph-os &                                   # KG MCP server (zero external services)
agent-utilities-doctor                       # sweep & verify the install

That's a working, durable-on-disk knowledge graph + MCP server with no database or external service (the tiny profile uses an in-process engine + embedded LadybugDB). Scale up later by re-running setup-config generate --profile single-node-prod.

1. Install¶

pip install agent-utilities[all]
# or a narrower set, e.g.:  pip install agent-utilities[owl,postgres,stardog]

2. Generate your config (all options)¶

Don't hand-write config.json — generate a complete, profile-seeded one that covers every option at a sensible default:

setup-config generate --profile tiny                 # → ~/.config/agent-utilities/config.json
setup-config reference                               # browse every option, grouped by subsystem

Profiles: tiny (laptop/edge), single-node-prod (one durable host), enterprise (multi-node). Secret-like keys are blanked — fill them via env or vault:// refs, never in the committed file.

3. (Optional) Databases — single-node-prod / enterprise¶

The tiny profile needs nothing here. For a durable Postgres tier (Apache AGE + pgvector + ParadeDB) and/or Stardog, run:

docker compose -f docker/pg-age-full.compose.yml up -d --build   # AGE + pgvector + pg_search
setup-databases --profile dev --dsn postgresql://agent:agent@localhost:5432/agent_kg

Full detail (prod Stardog, dev local SPARQL, backfill into AGE, OpenBao): databases recipe / the database-environment-setup skill.

4. Launch¶

graph-os                       # KG MCP server (stdio / streamable-http)
graph-os-daemon                # REST gateway (mounts /api/graph/*, /api/sparql, /metrics)
mcp-multiplexer                # one endpoint over the whole *-mcp fleet
# …or the interactive agent:
python -m agent_utilities --provider openai --model-id gpt-4o

5. Verify¶

Run the doctor — one sweep across config, engine, backend, secrets, auth, MCP fleet, hooks, and observability, each line carrying a fix + the skill that resolves it:

agent-utilities-doctor          # human-readable; --json for machines, --fix for safe auto-remediation, --live to probe endpoints

A HEALTHY (or WARNINGS) verdict + a graph_write/graph_query round-trip means you're up.

Use it¶

from agent_utilities import create_agent, create_agent_server

# Quick agent (skill_types selects which skill bundles to load)
agent = create_agent(name="MyAgent", skill_types=["universal", "graphs"])

# Full server with protocols (ACP, A2A, MCP, AG-UI)
create_agent_server(provider="openai", model_id="gpt-4o", port=8000)

See creating-an-agent.md for the complete agent walkthrough.

Console scripts (CLI reference)¶

Installed by the package:

Command	What it does
`setup-config {generate,doctor,reference}`	Generate the complete config.json, validate it, or list every option by subsystem
`setup-databases`	Provision Stardog + pg-age and backfill the graph into Apache AGE
`agent-utilities-doctor`	Holistic deployment health sweep (`--fix`, `--live`, `--json`)
`graph-os`	The Knowledge-Graph MCP server (graph-os)
`graph-os-daemon`	The REST gateway / KG daemon (`--status`)
`mcp-multiplexer`	Unified MCP tool gateway over the connector fleet
`agent-utilities-memory`	Memory store CLI
`python -m agent_utilities`	Launch the interactive agent (flags: `--provider`, `--model-id`, `--mcp-config`, `--web`, `--port`)

Each command is also reachable over MCP/REST via the graph_configure tool (generate_config, config_doctor, system_doctor, setup_databases, …).