UiPath Documentation
uipath-cli
latest
false
UiPath CLI user guide

uip rpa run, debug & execution

Syntax and options for `uip rpa run`, the `debug` group, and `execution cancel` — the current primary surface for running and debugging a workflow or coded file, replacing the older `run-file`/`stop-execution` verbs.

uip rpa run, uip rpa debug *, and uip rpa execution cancel run and debug a workflow or coded file through Studio. All 17 verbs on this page dispatch into the same two underlying tools that uip rpa run-file and uip rpa stop-execution expose as one wide --command enum — this is the current, primary, discoverable form of that same functionality, split into one verb per behavior instead of one flag with 19 values. run-file/stop-execution still work (they're kept as hidden compatibility aliases for installed skills), but this page is what --help actually shows.

Execution happens through Studio, so these commands work only on Windows runners (Studio is Windows-only). If Studio is not already running, it's brought up automatically — run uip rpa start-studio first if you want the startup cost paid in a dedicated setup step.

Synopsis​

uip rpa run --file-path <path> [--input-arguments <json>] [--log-level <level>] [--skip-build] [--profiling] [--profiling-mode endOfRun|stream] [--remote]

uip rpa debug start --file-path <path> [same options as run] [--breakpoints <json>]
uip rpa debug start-from-here [--input-arguments <json>] [--input-variables <json>] [--log-level <level>] [--profiling] [--profiling-mode <mode>]
uip rpa debug test-activity [--input-arguments <json>] [--input-variables <json>] [--log-level <level>] [--profiling] [--profiling-mode <mode>]
uip rpa debug break [--wait-timeout-seconds <n>]
uip rpa debug continue [--wait-timeout-seconds <n>]
uip rpa debug resume [--wait-timeout-seconds <n>]
uip rpa debug continue-retry [--wait-timeout-seconds <n>]
uip rpa debug continue-ignore [--wait-timeout-seconds <n>]
uip rpa debug step-into [--wait-timeout-seconds <n>]
uip rpa debug step-over [--wait-timeout-seconds <n>]
uip rpa debug step-out [--wait-timeout-seconds <n>]
uip rpa debug state [--wait-timeout-seconds <n>]
uip rpa debug set-breakpoints --breakpoints <json>
uip rpa debug apply-file-changes --file-changes <json>
uip rpa debug toggle-breakpoint
uip rpa debug restart-from-top

uip rpa execution cancel
uip rpa run --file-path <path> [--input-arguments <json>] [--log-level <level>] [--skip-build] [--profiling] [--profiling-mode endOfRun|stream] [--remote]

uip rpa debug start --file-path <path> [same options as run] [--breakpoints <json>]
uip rpa debug start-from-here [--input-arguments <json>] [--input-variables <json>] [--log-level <level>] [--profiling] [--profiling-mode <mode>]
uip rpa debug test-activity [--input-arguments <json>] [--input-variables <json>] [--log-level <level>] [--profiling] [--profiling-mode <mode>]
uip rpa debug break [--wait-timeout-seconds <n>]
uip rpa debug continue [--wait-timeout-seconds <n>]
uip rpa debug resume [--wait-timeout-seconds <n>]
uip rpa debug continue-retry [--wait-timeout-seconds <n>]
uip rpa debug continue-ignore [--wait-timeout-seconds <n>]
uip rpa debug step-into [--wait-timeout-seconds <n>]
uip rpa debug step-over [--wait-timeout-seconds <n>]
uip rpa debug step-out [--wait-timeout-seconds <n>]
uip rpa debug state [--wait-timeout-seconds <n>]
uip rpa debug set-breakpoints --breakpoints <json>
uip rpa debug apply-file-changes --file-changes <json>
uip rpa debug toggle-breakpoint
uip rpa debug restart-from-top

uip rpa execution cancel

Concepts​

  • One of two backends. Studio Desktop and the headless Helm backend both implement these; several options (--profiling-mode stream, debug set-breakpoints, debug apply-file-changes, and long-poll --wait-timeout-seconds) are Helm-only and are ignored or unavailable on the Studio backend — noted per-verb below.
  • --remote (on run/debug start/debug start-from-here/debug test-activity only) runs on the configured remote agent (an Orchestrator unattended robot) instead of locally — see uip rpa remote for how to configure the target first.
  • Every mid-session command returns at the next stable state. continue/resume/continue-retry/continue-ignore/the step verbs/break/state all wait for the session to reach a stable state (Paused, Suspended, or Completed) before returning; if none is reached within --wait-timeout-seconds they return DebugState: "Running" instead of hanging.

uip rpa run​

Run a workflow or coded file via Studio, with no debugger attached.

Options​

LongValueDescription
--file-path <path>stringRequired. Path to the .xaml workflow or .cs coded file.
--input-arguments <json>JSON objectProject-level input arguments as plain JSON, e.g. {"customerId":"ACME-1234"}.
--log-level <level>Verbose|Trace|Information|Warning|Error|CriticalMinimum log level in the output stream. Default Trace.
--skip-buildflagSkip validation/build, assuming the project already built. Use for rapid re-execution when nothing changed.
--profilingflagCollect per-activity profiling data, returned as a Profiling field. Requires a Studio Develop profile with EnableProfiling set — if unavailable, the run still completes and Profiling is simply omitted. Default false.
--profiling-mode <mode>endOfRun|streamDelivery timing for profiling data. endOfRun (default) returns the summary once at completion; stream also turns collection on by itself and additionally forwards each per-activity entry live as a [Profiling] line, plus writes the exhaustive data to the run's .uistat file. Helm backend only; ignored when running remote.
--remoteflagRun on the configured remote agent instead of locally. Configure the target first with uip rpa remote configure. Default false.

Example​

uip rpa run --file-path ./Main.xaml --input-arguments '{"customerId": "ACME-1234"}'
uip rpa run --file-path ./Main.xaml --input-arguments '{"customerId": "ACME-1234"}'

Data shape (--output json)​

{ "Code": "RunResult", "Data": { "DebugState": "Completed", "Result": {} } }
{ "Code": "RunResult", "Data": { "DebugState": "Completed", "Result": {} } }

Exact Data shape is the underlying tool's run-result payload — see uip rpa run-file for the full result contract (both verbs return the same shape).

uip rpa debug start​

Start a debug session for a workflow or coded file. Same options as run, plus --breakpoints.

Options​

All of run's options, plus:

LongValueDescription
--breakpoints <json>JSON arrayInitial breakpoint set. Each entry: workflowFile (required, path relative to project root), activityIdRef (the activity's sap2010:WorkflowViewState.IdRef from the XAML — the stable way to address it) or activityId (runtime id), plus optional condition (VB/C# expression; breaks only when true), hitCount (break only on the Nth hit), enabled (default true). Repeatable comma-separated key=value items, one array entry per occurrence, or a JSON file via @breakpoints.json / --breakpoints-file.

Example​

uip rpa debug start --file-path ./Main.xaml \
  --breakpoints 'workflowFile=Main.xaml,activityIdRef=Assign_1'
uip rpa debug start --file-path ./Main.xaml \
  --breakpoints 'workflowFile=Main.xaml,activityIdRef=Assign_1'

Data shape (--output json)​

Same shape as run; a session that hits a breakpoint reports DebugState: "Paused" with the activity and locals in DebugDetails.

uip rpa debug start-from-here​

Start debugging from the currently-focused activity, skipping everything before it. No --file-path — it operates on whatever's focused in the open project.

Options​

LongValueDescription
--input-arguments <json>JSONProject-level inputs. Unlike run/debug start, values here are VB.NET or C# expressions, not plain JSON literals.
--input-variables <json>JSONWorkflow-level variable values to seed in-scope state (expressions, not literals).
--log-level <level>same enum as runMinimum log level.
--profiling / --profiling-mode <mode>—Same as run.

Example​

uip rpa debug start-from-here --input-variables '{"greeting": "\"Hello World\""}'
uip rpa debug start-from-here --input-variables '{"greeting": "\"Hello World\""}'

uip rpa debug test-activity​

Isolate and execute the currently-focused activity, using --input-arguments/--input-variables (expressions, same convention as start-from-here) to seed in-scope state.

Options​

Same as debug start-from-here.

Example​

uip rpa debug test-activity --input-variables '{"greeting": "\"Hello World\""}'
uip rpa debug test-activity --input-variables '{"greeting": "\"Hello World\""}'

uip rpa debug break​

Pause execution at the next executed activity in the active session.

Options​

LongValueDescription
--wait-timeout-seconds <n>numberMax seconds to wait for the next stable state before returning DebugState: "Running". Default 120. Helm backend only — the Studio backend applies its own fixed 120s bound and ignores an explicit value.

uip rpa debug continue / resume / continue-retry / continue-ignore​

Resume execution in the active session. continue resumes from paused; resume resumes from suspended; continue-retry resumes and retries the activity that just threw; continue-ignore resumes and swallows the exception from the activity that just threw.

Options​

Same --wait-timeout-seconds as debug break.

uip rpa debug step-into / step-over / step-out​

Standard debugger stepping in the active session. step-out, called with no active session, starts one paused before the first activity.

Options​

Same --wait-timeout-seconds as debug break.

uip rpa debug state​

Report the active debug session's state (Running/Paused/Suspended/Completed/None) with no side effects.

Options​

LongValueDescription
--wait-timeout-seconds <n>numberLong-poll a running session for its next stable state. Default 0 (an instant probe). Helm backend only — on the Studio backend this is always an instant probe.

uip rpa debug set-breakpoints​

Replace the active debug session's whole breakpoint set. Helm backend only — on the Studio backend, use debug toggle-breakpoint, or pass breakpoints to debug start instead.

Options​

LongValueRequiredDescription
--breakpoints <json>JSON arrayyesSame shape as debug start's --breakpoints. Replaces the whole set, not a merge.

uip rpa debug apply-file-changes​

While paused/suspended in a debug session, reconcile on-disk activity-property edits into the running session via hot-reload, so it picks them up without a restart. Edit the workflow file first, then call this naming what changed.

Options​

LongValueRequiredDescription
--file-changes <json>JSON arrayyesOne object per edited property: workflowFile (required), activityIdRef, plus the changed property name/value. Repeatable key=value form or @changes.json / --file-changes-file.

Data shape (--output json)​

{ "Code": "ApplyFileChanges", "Data": { "applied": [], "rejected": [], "debugState": "Paused" } }
{ "Code": "ApplyFileChanges", "Data": { "applied": [], "rejected": [], "debugState": "Paused" } }

applied edits are live in the session (e.g. continue-retry re-runs a faulted activity with the fix); rejected edits (each with a reason) are structural changes that need a session restart to take effect.

uip rpa debug toggle-breakpoint​

Toggle a breakpoint at the currently focused activity (.xaml) or line (.cs). No options. XAML cycles enabled → disabled → none; coded workflows cycle on/off.

uip rpa debug restart-from-top​

Restart the active debug session from the beginning. No options.

uip rpa execution cancel​

Cancel the currently active execution or debug session — works uniformly for both run and debug start. No options.

Example​

uip rpa execution cancel
uip rpa execution cancel

See also​

Was this page helpful?

Connect

Need help? Support

Want to learn? UiPath Academy

Have questions? UiPath Forum

Stay updated