feat: add suspend/resume support for RPA invocations in evaluations #1083

Chibionos · 2026-01-09T22:26:28Z

Summary

Adds support for suspend/resume pattern in evaluation runtime, enabling agents to suspend execution when invoking RPA processes and resume after completion.

Key Changes

Suspend Detection (_runtime.py)

Runtime detects UiPathRuntimeStatus.SUSPENDED after agent execution
Extracts trigger information from suspended executions
Skips evaluators during suspension (evaluators only run on SUCCESSFUL completion)
Returns triggers in partial results for resume context

Resume Support (cli_eval.py)

Added --resume flag to evaluation CLI
Loads checkpoint state and continues from interruption point
Passes resume context to runtime

Trigger Pass-Through

Suspended executions return triggers with structure:
- interruptId: Unique interrupt identifier
- triggerType: Type of trigger (e.g., "Api")
- payload: InvokeProcess details (process name, input arguments, folder path)

Comprehensive Logging

Suspension detection events (🔴 DETECTED SUSPENSION)
Trigger extraction details (📋 Extracted N trigger(s))
Evaluator skip reasons (⏭️ Skipping evaluators)
Trigger pass-through confirmation (✅ Passing through triggers)
Resume mode indication (🟢 RESUMED execution)

Testing

Testing is done via the sample agent in the companion PR: UiPath/uipath-langchain-python#397

Quick Testing Guide

Install the Dev Version

pip install git+https://github.com/UiPath/[email protected]

Run Quick Test

cd samples/event-trigger

# Test 1: Direct runtime test
python -c "
import asyncio
from uipath.functions.runtime import UiPathFunctionsRuntime

async def test():
    runtime = UiPathFunctionsRuntime(
        'test_suspend_resume_agent.py',
        'main', 
        'test_suspend_resume_agent'
    )
    result = await runtime.execute({'query': 'Test'})
    print(f'✓ Status: {result.status}')
    print(f'✓ Has Trigger: {result.trigger is not None}')
    if result.trigger:
        print(f'✓ Process: {result.trigger.payload[\"process_name\"]}')

asyncio.run(test())
"

# Test 2: Evaluation test
uipath eval test_suspend_resume_agent evaluations/test_suspend_resume.json --output-file test_output.json
cat test_output.json | python -m json.tool | grep -A 5 "status"

Expected Results

✅ Status: SUSPENDED
✅ Trigger exists with process name and arguments
✅ Evaluation output shows suspended state in JSON

Key Files Changed

Functions Runtime: src/uipath/functions/runtime.py (+70 lines)
- New _detect_langgraph_interrupt() method
- Trigger extraction from __interrupt__ field
Evaluation Runtime: src/uipath/_cli/_evals/_runtime.py (+58 lines)
- SUSPENDED status detection
- Skip evaluators when suspended
- Publish suspended event
Test Agent: samples/event-trigger/test_suspend_resume_agent.py (new file)
- Demonstrates correct interrupt pattern
- Uses checkpointer and returns raw dict

Tag: v2.4.9-dev.suspend-resume
Branch: investigate-rpa-samples

This change implements the suspend/resume pattern for RPA process invocations within the evaluation runtime, ensuring evaluations can pause when waiting for external job completion and resume after the job finishes. Key Changes: 1. Functions Runtime (src/uipath/functions/runtime.py): - Added _detect_langgraph_interrupt() method to detect LangGraph's __interrupt__ field and extract trigger information - Modified execute() to check for interrupts and return SUSPENDED status - Converts InvokeProcess objects to UiPathResumeTrigger for serverless executor 2. Evaluation Runtime (src/uipath/_cli/_evals/_runtime.py): - Added SUSPENDED status detection after agent execution - Populates agentExecutionOutput with trigger data when suspended - Skips evaluator execution for suspended runs - Publishes EvalRunUpdatedEvent with suspended status to event bus 3. Test Agent (samples/event-trigger/test_suspend_resume_agent.py): - Updated to use MemorySaver checkpointer (required for LangGraph interrupts) - Returns raw dict to preserve __interrupt__ field for runtime detection The implementation follows the established pattern from UiPathResumableRuntime and ensures proper trigger extraction and status handling throughout the evaluation lifecycle. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <[email protected]>

This commit adds three key features to support suspend/resume in evaluations: 1. Add --resume option to uipath eval command - Added resume flag to CLI arguments - Passed through to UiPathEvalContext - Enables resuming suspended evaluations 2. Use known execution_id for single runtime runs - Added job_id field to UiPathEvalContext - Passed ctx.job_id from UiPathRuntimeContext to eval context - UiPathEvalRuntime now uses job_id as execution_id when available - Falls back to UUID generation when job_id is None 3. Return inner runtime triggers to eval runtime result - Collect all triggers from evaluation run results - Pass triggers through to top-level UiPathRuntimeResult - Enables serverless executor to detect and process suspend triggers - Preserves trigger information from nested runtime executions These changes work with the suspend/resume detection already implemented to enable full suspend/resume flow for RPA process invocations in evaluations. Testing will be done with the test agent in uipath-langchain-python repo.

…runtime Added detailed logging to track the suspend/resume flow: Execution Start: - Log execution ID, job ID, and resume mode status - Clear indication when resume mode is enabled Suspension Detection: - Log when SUSPENDED status is detected - Show number of triggers extracted - Display each trigger's details Trigger Pass-through: - Log trigger collection from all evaluation runs - Show count and details of triggers being passed through - Clear confirmation when triggers reach top-level result This makes the suspend/resume mechanism transparent and easy to debug for both development and production troubleshooting.

The draw.io diagram is not working properly. Testing will be done in the uipath-langchain-python repo sample instead.

mjnovice · 2026-01-12T18:01:37Z

Just to understand, so the suspend and resume is actually being done by the agent and not the orchestrator, right ?

cristipufu · 2026-01-14T11:08:35Z

src/uipath/functions/runtime.py

 from uipath.runtime.errors import (
    UiPathErrorCategory,
    UiPathErrorCode,
    UiPathErrorContract,


you need to undo everything in this file (this is used for arbitrary python code, no langgraph deps)

cristipufu · 2026-01-14T11:09:10Z

src/uipath/_cli/_evals/_runtime.py

+
                        result = UiPathRuntimeResult(
                            output={**results.model_dump(by_alias=True)},
                            status=UiPathRuntimeStatus.SUCCESSFUL,


if the runtime is suspended, you need to pass the result as suspended to the serverless exector

github-actions bot added test:uipath-langchain Triggers tests in the uipath-langchain-python repository test:uipath-llamaindex Triggers tests in the uipath-llamaindex-python repository labels Jan 9, 2026

Chibionos changed the title ~~feat: Add suspend/resume support for RPA invocations in evaluations~~ feat: add suspend/resume support for RPA invocations in evaluations Jan 10, 2026

Chibi Vikram and others added 6 commits January 12, 2026 07:26

docs: add suspend/resume eval runtime architecture diagram

617ffba

chore: remove draw.io diagram and event-trigger test agent

3722a69

The draw.io diagram is not working properly. Testing will be done in the uipath-langchain-python repo sample instead.

style: fix linting issues in functions runtime

a35d2a5

Chibionos force-pushed the investigate-rpa-samples branch from 0a124f0 to a35d2a5 Compare January 12, 2026 15:26

cristipufu reviewed Jan 14, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: add suspend/resume support for RPA invocations in evaluations #1083

feat: add suspend/resume support for RPA invocations in evaluations #1083

Chibionos commented Jan 9, 2026 •

edited

Loading

Uh oh!

Chibionos commented Jan 9, 2026

Uh oh!

mjnovice commented Jan 12, 2026

Uh oh!

cristipufu Jan 14, 2026

Uh oh!

cristipufu Jan 14, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

feat: add suspend/resume support for RPA invocations in evaluations #1083

Are you sure you want to change the base?

feat: add suspend/resume support for RPA invocations in evaluations #1083

Conversation

Chibionos commented Jan 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Key Changes

Testing

Related

Uh oh!

Chibionos commented Jan 9, 2026

Quick Testing Guide

Install the Dev Version

Run Quick Test

Expected Results

Key Files Changed

Uh oh!

mjnovice commented Jan 12, 2026

Uh oh!

cristipufu Jan 14, 2026

Choose a reason for hiding this comment

Uh oh!

cristipufu Jan 14, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Chibionos commented Jan 9, 2026 •

edited

Loading