paseo/public-docs/cli.md

title, description, nav, order, category

title	description	nav	order	category
CLI reference	Paseo CLI reference: manage projects, workspaces, agents, plugins, scripts, schedules, daemons, and permissions from your terminal.	CLI reference	35	Orchestration

CLI reference

The Paseo CLI lets you manage agents from your terminal. It's the same interface exposed by the daemon's API, so anything you can do in the app you can do from the command line.

Agent orchestration: You can tell coding agents to use the Paseo CLI to spawn and manage other agents. Paseo recognizes the calling agent, so CLI-created workers get the same workspace and parent defaults as MCP-created workers.

Quick reference

paseo run "fix the tests"            # Start an agent
paseo ls                             # List running agents
paseo attach <id>                    # Stream agent output
paseo send <id> "also fix linting"   # Send follow-up task
paseo logs <id>                      # View agent timeline
paseo stop <id>                      # Stop an agent
paseo usage                          # Show account quotas and balances

Account usage

Show account quotas and balances from the host's usage sources:

paseo usage
paseo usage ls --refresh
paseo usage ls --agent <agent-id>
paseo usage inspect <report-id>
paseo usage inspect <report-id> --refresh

paseo usage is equivalent to paseo usage ls. The summary groups windows and balances by account, with report IDs, source and account labels, status, and fetch timestamps. Missing values appear as -; unavailable and failed reports include their reasons.

Copy the full report ID from the list into inspect to see reset times, balances and units, plan details, supplied forecasts, and login failures with recovery instructions. Report IDs require exact matches.

--agent selects the account quota associated with an agent. It does not measure that agent's token consumption or cost. Use the full agent ID, and do not combine --agent with inspect.

Queries use the host's five-minute cache when available. Add --refresh to bypass it. Use --host or --home to select a daemon. Hosts without usage-source support return an update message.

Usage in scripts

paseo usage ls --json
paseo usage inspect <report-id> --json
paseo usage ls --quiet

List JSON is always an array, including [] when no reports are returned. Inspect JSON is one report entry. Each entry includes id, sourceId, sourceLabel, account, fetchedAt, and a nested report. Read report.status before accessing its data:

Status	Fields
available	windows, with optional planLabel, balances, and details
unavailable	problem, describing expired or rejected credentials or missing quota
error	error, the failure message

Entries may also include loginErrors. Structured output preserves numeric values and optional fields. Standard --format yaml, --no-headers, and --no-color options are supported; --quiet prints full report IDs.

A completed query exits with status 0, including when individual reports are unavailable or failed. Connection or request failures, unsupported hosts, unknown report IDs, and conflicting selectors exit with status 1 and write the error to stderr. Human output says No usage reports returned for an empty result; this does not establish whether account discovery succeeded for every source.

Provider diagnostics

Ask the daemon to inspect the provider environment it actually uses:

paseo provider diagnostic claude
paseo provider diagnostic codex --json
paseo --host devbox:6767 provider diagnostic opencode

The diagnostic includes the configured command, daemon PATH and shell, matching binaries, resolved path, version, model count, and provider status. Use the global --host option for a remote daemon. This is the same diagnostic shown under Settings → your host → Providers → provider → Diagnostic.

Running agents

Use paseo run to start a new agent with a task:

paseo run "implement user authentication"
paseo run --provider codex "refactor the API layer"
paseo run --no-wait "run the focused test suite"
paseo run --output-schema schema.json "name this branch"
paseo run --new-workspace worktree --worktree-mode branch-off --new-branch feature/x --base origin/main "implement feature X"
paseo run --workspace <workspace-id> "review the current diff"
paseo run --output-schema schema.json "extract release notes"
paseo run --output-schema '{"type":"object","properties":{"summary":{"type":"string"}},"required":["summary"]}' "summarize release notes"

From a human shell, a bare paseo run creates a new local workspace for the current directory. Use --workspace <id> to add the agent to an existing workspace, or --new-workspace local|worktree to explicitly create a separate workspace for the run.

Worktree creation accepts --worktree-mode branch-off|checkout-branch|checkout-pr plus the matching --new-branch/--base, --branch, or --pr-number/--forge options. Use --worktree-slug to choose the managed directory slug.

When an existing Paseo agent runs the same command, Paseo recognizes it through PASEO_AGENT_ID. Without explicit placement, the new agent becomes its subagent in the same workspace. --workspace can place that subagent elsewhere without changing its parent.

Use --output-schema to return only matching JSON output. You can pass a schema file path or an inline JSON schema object. This mode cannot be used with --no-wait.

By default, paseo run waits for completion. Use --no-wait to return immediately while the agent keeps running, the same flag paseo send takes.

The old run --background and -d aliases still mean --no-wait. They are deprecated, hidden from help, and print a warning to stderr. Use --no-wait in scripts. Workspace visibility is selected separately with workspace create --background; this release does not repurpose run --background.

Projects

Register the current directory as a project, then list the projects known to the daemon:

cd ~/dev/my-app
paseo project create
paseo project ls

Use the project ID from paseo project ls to rename, reset, or delete a project:

paseo project rename <project-id> "My app"
paseo project rename <project-id> --reset
paseo project delete <project-id>

--reset restores the name derived from the project directory. Deleting a project archives its active workspaces and removes the project from Paseo. It does not delete the project directory.

For a local daemon, paseo project create [path] defaults to the current directory and resolves relative paths on the CLI machine. When you use the global --host option or PASEO_HOST, provide a path that the target daemon can access:

paseo --host devbox:6767 project create /srv/repos/api

The remote daemon interprets that path on its own machine. See Workspaces for how projects group working directories and sessions.

Workspaces

Create a workspace independently when you want to prepare its files before starting an agent:

paseo workspace create --isolation local --path ~/dev/my-app --title main

paseo workspace create \
  --isolation worktree \
  --path ~/dev/my-app \
  --mode branch-off \
  --new-branch feature/auth \
  --worktree-slug feature-auth \
  --base origin/main

paseo workspace create \
  --isolation worktree \
  --path ~/dev/my-app \
  --mode checkout-branch \
  --branch feature/existing \
  --worktree-slug existing-copy

paseo workspace create \
  --isolation worktree \
  --path ~/dev/my-app \
  --mode checkout-pr \
  --pr-number 2186

Add --background to hide a workspace from default discovery. Its agents and terminals keep ordinary history, persistence, hooks, and restart recovery. --no-background explicitly overrides caller-workspace inheritance; omission inherits when run by a Paseo agent and defaults to false from a human shell. Visibility is creation-only.

paseo workspace create --isolation local --background --path /path/to/repo
paseo run --workspace <workspace-id> --no-wait "review the checkout flow"

Exact workspace and agent IDs remain accessible regardless of listing visibility.

Then list, use, rename, or archive it:

paseo workspace ls
paseo workspace ls --background               # Also include background workspaces
paseo run --workspace <workspace-id> "implement authentication"
paseo workspace rename <workspace-id> "Auth rework"
paseo workspace rename <workspace-id> --reset   # back to the branch or directory name
paseo workspace archive <workspace-id>

Add --forge <name> to PR checkout when Paseo cannot infer the forge from the source checkout. See Git worktrees for setup hooks and services.

Terminals

Use the workspace ID when multiple workspaces share a directory:

paseo terminal create --workspace <workspace-id> --name Development
paseo terminal ls --workspace <workspace-id> --json
paseo terminal send-keys <terminal-id> -l "echo ready"
paseo terminal send-keys <terminal-id> Enter
paseo terminal capture <terminal-id>
paseo terminal kill <terminal-id>

Creation defaults to the workspace directory. Add --cwd <absolute-path> to change the process directory while keeping that workspace as the owner. Unknown and archived workspace IDs fail.

Without --workspace, creation opens the project at --cwd or the current directory and reuses its oldest active workspace. Listing without --workspace filters by --cwd or the current directory and can include multiple workspaces. ls --all lists every terminal on the host and cannot be combined with directory or workspace filters.

Create and list results include id, name, cwd, and workspaceId. Use --json for structured output and the global --host option to target another daemon. These commands require a host that supports the workspace terminal API; older hosts return an update message.

Workspace scripts

List, start, and stop the scripts configured in a workspace's paseo.json:

paseo script ls
paseo script start web
paseo script stop web

By default, Paseo selects the workspace whose directory is the current directory. Pass --cwd <path> to select a different directory, or --workspace <workspace-id> when a directory has multiple workspaces. Use the global --host option to target another daemon. These commands also accept standard output options such as --json.

The output includes each script's lifecycle and supervised terminal ID. Services also include their assigned port, proxy URL, and health. See Git worktrees for paseo.json configuration.

Plugins

Trust every plugin you add. paseo plugin add and paseo plugin install mean “I trust this codebase.” Plugin server code and Git preparation commands run unsandboxed with the daemon user's access on the daemon host; client contributions run inside Paseo. Dependencies and future updates are part of that decision. With the global --host option, commands run on the remote daemon host.

Create and manage trusted plugins on a daemon:

paseo plugin init ./plugin
paseo plugin install ./plugin
paseo plugin add owner/slug
paseo plugin add git:owner/repository
paseo plugin add https://gitlab.com/group/repository.git --ref main
paseo plugin add git:owner/monorepo:plugins/review
paseo plugin ls [id]
paseo plugin update my-plugin
paseo plugin update --all
paseo plugin reload my-plugin
paseo plugin logs my-plugin
paseo plugin disable my-plugin
paseo plugin enable my-plugin
paseo plugin remove my-plugin

paseo plugin reload <id> reports the result of the reload itself: it waits while the daemon stops and starts the plugin, however long that takes, and it queues behind the daemon's other plugin work. Cancelling the command or losing the connection does not stop the daemon's work, and no listing or log entry settles what a cancelled reload did — the Plugin reference covers those interim states. Running the reload again is what reports the result: it waits behind the outstanding one, then stops and starts the plugin once more.

Bare owner/slug installs a reviewed registry listing. Use git:owner/repository for GitHub source, or ./plugin for a local directory. Append :<directory> for a plugin in a monorepo.

paseo plugin ls [id] does not contact the remote. paseo plugin logs <id> returns the plugin's recent daemon-side stdout and stderr. Add --json for structured entries, or run paseo --host <target> plugin logs <id> for another daemon. See the Plugin reference for installation, trust, lifecycle, and log-retention behavior.

Listing agents

paseo ls                    # Non-archived agents in active workspaces
paseo ls -a                 # Also include archived agents
paseo ls -g                 # Non-archived agents across all workspaces
paseo ls -a -g --json       # All agents, including archived, as JSON
paseo ls --background       # Also include agents in background workspaces

Streaming output

Use paseo attach to stream an agent's output in real-time:

paseo attach abc123   # Attach to agent (Ctrl+C to detach)

Agent IDs can be shortened, abc works if it's unambiguous.

Sending messages

Send follow-up tasks to a running or idle agent:

Use the recipient's agent ID from paseo ls, or copy it from the agent's tab.

paseo send <id> "now run the tests"
paseo send <id> --image screenshot.png "what's wrong here?"
paseo send <id> --no-wait "queue this task"

Viewing logs

paseo logs <id>                  # Full timeline
paseo logs <id> -f               # Follow (streaming)
paseo logs <id> --tail 10        # Last 10 entries
paseo logs <id> --filter tools   # Only tool calls

Waiting for agents

Block until an agent finishes its current task:

paseo wait <id>
paseo wait <id> --timeout 60   # 60 second timeout

Useful in scripts or when one agent needs to wait for another.

Schedules

Run an agent on a cron schedule. The CLI also accepts simple cadence presets and compiles them to cron. See Schedules from the CLI for the full reference.

paseo schedule create --every 30m --cwd ~/dev/my-app "Continue the refactor and leave a note."
paseo schedule ls
paseo schedule pause <id>

Permissions

Agents may request permission for certain actions. Manage these from the CLI:

paseo permit ls                # List pending requests
paseo permit allow <id>        # Allow all pending for agent
paseo permit deny <id> --all   # Deny all pending

Agent modes

Change an agent's operational mode (provider-specific):

paseo agent mode <id> --list   # Show available modes
paseo agent mode <id> bypass   # Set bypass mode
paseo agent mode <id> plan     # Set plan mode
paseo agent detach <id>        # Make a subagent top-level

Detaching is an explicit lifecycle action, not a creation flag. The agent keeps running; only its relationship to its parent changes.

Daemon management

Define an instance once, then start its saved configuration:

paseo daemon config set daemon.listen 127.0.0.1:6799 --home ~/paseo-test
paseo daemon config set daemon.relay.enabled false --home ~/paseo-test
paseo daemon start --home ~/paseo-test
paseo project ls --home ~/paseo-test
paseo daemon restart --home ~/paseo-test
paseo daemon stop --home ~/paseo-test

start runs in the background and reports the actual listening address and supervisor PID. It accepts only home selection and --timeout <seconds> (default 600). If waiting times out, the supervisor remains running: use the printed status, log, and stop instructions. A worker that exits before becoming ready makes startup fail.

restart requests a replacement worker from the existing supervisor. It rereads the file and retains the supervisor's original environment and arguments. Success confirms a different ready worker, following a changed address for a home target. It never starts a stopped daemon or refreshes the supervisor binary. A timeout reports whether the request was acknowledged; it does not prove why reconnection failed.

stop --home waits for that local supervisor to exit. On POSIX it signals the supervisor without contacting a TCP endpoint. On Windows it uses the ready daemon's shutdown RPC; an unbound instance requires explicit --force. --force permits forced process-tree cleanup after the graceful timeout (default 15 seconds). stop --host only reports shutdown requested; remote process exit is not verified. A service manager may start another instance after the captured supervisor exits.

status separates local supervisor state, its published endpoint, the configured address, and RPC reachability. A stopped home is never probed at its configured address. An unbound live supervisor is not ready. If an authenticated local connection remains open but status details time out, the result stays reachable with a note explaining the unavailable details. Worker and provider fields are omitted. An explicit --host query still fails when its status request fails.

reload validates the file, applies runtime-safe changes, and reports appliedPaths, restartRequiredPaths, and overrideControlledPaths. It never implicitly restarts. Use --json or --format yaml for structured results. An older host lacking the capability reports that it needs an update.

The root aliases start, status, restart, reload, and pair use the same commands as daemon. Root run and stop remain agent operations.

Foreground deployments and migration

Use environment overrides with the foreground deployment command:

PASEO_LISTEN=127.0.0.1:6799 PASEO_RELAY_ENABLED=false paseo daemon run --home ~/paseo-test

It stays attached until the supervisor exits or you cancel, without a readiness timeout. Worker restart retains these launch inputs. Stop and relaunch the deployment to change them. If the home already has a live supervisor, run returns already_running without owning or launching a foreground process.

Managed start ignores inherited daemon-setting overrides, including PORT, PASEO_LISTEN, relay, voice, and web UI settings. It preserves provider credentials and executable/runtime controls. start --foreground is removed; use daemon run. Former start/restart configuration flags such as --port, --no-relay, and --web-ui fail before side effects, with the corresponding config set migration. See configuration edits.

Select one daemon

Every daemon-connected CLI command accepts global --home or --host, before or after the command. A home selects a local supervisor's published endpoint; a host selects an explicit endpoint. There is no configured-address or default-port fallback.

Selectors	Result
--home	That local home, overriding both environment selectors
--host	That endpoint, overriding both environment selectors
Both flags, or conflicting duplicate flags	TARGET_AMBIGUOUS
Only PASEO_HOME or only PASEO_HOST	The corresponding target
Both environment selectors, without a flag	TARGET_AMBIGUOUS
Neither	Default local home, ~/.paseo

Local-only start, daemon run, config, onboard, and set-password reject explicit --host and ignore PASEO_HOST. Endpoint operations retain TCP, Unix socket, Windows pipe, SSH, and pairing-offer transports. A host-side CLI controlling a container needs --host or PASEO_HOST.

Hub

paseo hub login [url]          # Approve and store organization-scoped CLI access
paseo hub init                 # Create and optionally deploy a starter trigger here
paseo hub connect [url]        # Enroll this daemon using CLI access
paseo hub projects             # List legacy projects in the authenticated organization
paseo hub status               # Show the current Hub relationship
paseo hub permissions list     # Show what this Hub may do on this daemon
paseo hub permissions grant hub.execute    # Let Hub automations run agents here
paseo hub permissions revoke hub.execute   # Take it back
paseo hub disconnect           # End it
paseo hub deploy               # Validate and install .paseo/triggers/*.yml
paseo hub deploy --dry-run     # Validate without installing
paseo hub deploy -p <project>   # Deploy an existing legacy project bundle
paseo hub logout               # Remove the active stored CLI login

Run deploy from the repository root. By default it reads every direct .paseo/triggers/*.yml file in deterministic path order. It validates all triggers before installing them one at a time. If an installation fails after earlier ones succeeded, the error lists the installed files. --dry-run only validates; it does not create or activate revisions.

Pass -p, --project <slug> for an existing legacy bundle: .paseo/hub.yml, direct .paseo/workflows/*.yml files, and referenced workflow partials. See Deploy from the CLI.

login opens the Hub approval page and stores a durable organization-scoped CLI credential under PASEO_HOME. In an interactive terminal it offers to connect this daemon, then separately asks whether to allow Hub automations to run agents. Connection defaults to yes; execution permission defaults to no. It then links to Hub's Triggers page and prints paseo hub init for setup as code. --json and non-TTY login remain login-only and never prompt. The stored login is separate from the daemon relationship created by connect.

init requires a TTY. It signs in and connects the daemon as needed, then lists the organization's app connections that can back a starter trigger. One usable connection is selected automatically; with several, you choose a Trigger connection. If none is ready, setup sends you to Hub → Apps and stops before selecting an agent or writing files.

Setup asks which agent provider, model, and mode to run. Providers must be enabled and expose both a selectable model and an execution mode. Suggested model and mode entries are the daemon's defaults; a mode is still selected explicitly when there is no default. Hub validates the choice against the daemon before deploying, and accepts only Claude, Codex, and OpenCode for its unattended runs. deploy applies the same checks, so it needs the named daemon connected. Setup then asks for the identity allowed to trigger the bot: a GitHub username, Slack member ID, or Discord user ID. It validates the trigger, writes .paseo/triggers/<provider>-help.yml, and asks whether to deploy. Replacing that file requires confirmation; existing legacy bundles and other trigger files are preserved. See the generated starter trigger.

Interactive logout checks the same-origin daemon relationship and asks whether to disconnect before deleting the login. Declining removes only the login. JSON and noninteractive logout never prompt or disconnect implicitly; --disconnect-daemon is the explicit automation path, and --force applies to that daemon disconnection. If a requested disconnection fails, the login is preserved.

Every command resolves and normalizes its destination before Hub or daemon work. Origin precedence is an explicit command origin or --hub, then PASEO_HUB_URL, then the active stored login origin, then the hosted default https://hub.paseo.sh. The hosted default never overrides an active login. Credential precedence is --api-key <secret>, then PASEO_HUB_API_KEY, then a stored login for the exact resolved origin. A stored credential is never sent to a different origin. API keys passed through flags or the environment are not stored.

Human output reports the resolved destination before each action. JSON output keeps stdout machine-readable and includes the normalized Hub origin. Bundle diagnostics identify paths without printing configuration contents or credentials.

See Daemons in Hub, Hub configuration, and the Hub public API.

Connecting to a remote daemon

The global --host option accepts either a local target (host:port, a unix socket, or a Windows pipe) or a pairing offer URL, the same https://app.paseo.sh/#offer=... link the mobile app uses for QR pairing. With an offer URL the CLI connects through the Paseo relay with end-to-end encryption, so you can drive a daemon on another machine without exposing it to the network.

Get an offer URL from the daemon you want to control:

paseo daemon pair          # prints the QR and link when relay is enabled
paseo daemon pair --relay  # enables relay without prompting
paseo daemon pair --json   # structured output; never prompts

Relay is off for new installations. A disabled relay returns a RELAY_DISABLED error; pass --relay to provide explicit consent. For a stopped home, pairing is labelled offline; --relay saves relay enablement and the offer includes a start instruction. A live but unreachable home never falls back to an offline identity. Relay pairing is end-to-end encrypted. See Security.

Use it from anywhere:

paseo --host 'https://app.paseo.sh/#offer=eyJ2IjoyLC...' ls
paseo --host "$OFFER_URL" run "fix the failing tests"

You can also set it once via PASEO_HOST instead of passing --host on every command. An explicit flag overrides the environment variable.

Multi-agent workflows

The CLI is designed to be used by agents themselves. You can instruct an agent to spawn sub-agents for parallel work:

# Agent A spawns Agent B and waits for it
agent_id=$(paseo run --no-wait --quiet --title api-agent "implement the API")
paseo wait "$agent_id"
paseo logs "$agent_id" --tail 5

Because Agent A's ID is present in the environment, Agent B is created as its subagent in the same workspace unless --workspace is specified.

Simple implement + verify loop:

# Requires jq
while true; do
  paseo run --provider codex "make the tests pass" >/dev/null

  verdict=$(paseo run --provider claude --output-schema '{"type":"object","properties":{"criteria_met":{"type":"boolean"}},"required":["criteria_met"],"additionalProperties":false}' "ensure tests all pass")
  if echo "$verdict" | jq -e '.criteria_met == true' >/dev/null; then
    echo "criteria met"
    break
  fi
done

This pattern enables hierarchical task decomposition, a lead agent can break down work, delegate to specialists, and synthesize results.

Output formats

Most commands support multiple output formats for scripting:

paseo ls --json                # JSON output
paseo ls --format yaml         # YAML output
paseo ls -q                    # IDs only (quiet)

Global options

• --host <target>, connect to a different daemon (host:port, unix socket, or https://app.paseo.sh/#offer=... for relay). See Connecting to a remote daemon.
• --json, JSON output
• -q, --quiet, minimal output
• --no-color, disable colors