Quick start

Ten minutes, three surfaces. Nothing here needs configuration and nothing here writes to your machine.

1. Ask it something

rta sys cpu

model  Apple M3 Max
cores  14 physical, 14 logical
usage  62.5%

Every capability works the same way, and there are 101 of them in the default build:

rta sys overview            # grouped host health
rta net dns github.com      # DNS, with --type auto
rta cert check example.com  # TLS chain and expiry
rta fs usage ~/Downloads    # what is using space
rta git status              # a repository's state, structured
rta audit web example.com   # TLS, headers, cookies, exposure — graded

2. Get it in the shape you need

The same command renders five ways. --output (or -o) is on every command:

rta sys cpu -o json | jq '.rows'
rta net dns github.com -o csv
rta sys overview -o md >> report.md

pretty is the default when a human is looking. Scripts should say what they want.

3. Open the shell

rta

Bare rta on a terminal opens the interactive TUI — a live dashboard with a search bar over every capability, one tile per plugin, and forms for anything that takes input. In a pipe it prints help instead, so a script never hangs on an invisible TUI.

Press / to search, enter to run, esc to go back, q to quit.

4. Find out what anything does

rta explain sys.cpu

id           sys.cpu
summary      Show CPU model, core count and current usage
safety       read
idempotent   true
cli          rta sys cpu [--cores <bool>]
mcp-tool     sys_cpu
input:cores  bool — per-core usage as a bar chart

That card is not documentation about the capability — it is generated from the same declaration the CLI, the TUI and the MCP schema are built from, so it cannot drift. rta explain with no argument lists everything.

5. Connect an agent

This is the part worth slowing down for.

rta mcp install claude

That registers rta with Claude Code under the name claude. The name matters: grants you issue while talking to one client do not follow every other client on this machine.

For a client that has no command of its own, rta prints exactly what to add and where, and writes nothing. It will not edit another tool's config file — that file is what gives an agent access to your secrets, and it is worth reading before it changes.

rta mcp install cursor --show

What the agent can do now

Read-only. Everything else is refused. The agent can call sys.cpu, net.dns, git status and every other read capability. It cannot write, cannot delete, and cannot read your secret store.

That is the default with no configuration, no flags, and no decisions from you.

Letting it do one more thing

rta grant allow kv.get db-password --ttl 30m --max-uses 1

That allows one key, for thirty minutes, once. Not the store — that key. When any of those three bounds is reached, it stops.

rta grant list      # what is allowed right now
rta agent log       # what agents actually did, refusals included

6. Check what you have exposed

rta doctor

Read the info rows rather than skipping to the failures. Lines like "the store unlocks from this environment — an MCP server started here can read secrets, bounded only by grants" are the ones that tell you what an agent started from this shell inherits.

Where to go next

If you want to…	Read
Script rta, or use it in CI	The CLI
Understand what an agent can reach	MCP and the safety gate
Grant something narrowly	Grants
Store credentials	Secrets
Point rta at staging vs production	Profiles
Add postgres, S3, Vault, Kubernetes	Using plugins
See it all working together	Recipes