Fix: runflow toolmode input progation

[HzaRashid]

• Fixes Empty input when using RunFlow component #11153 by extracting nested flow_tweak_data from self._attributes.
• Fixes Run Flow Component can only be used as start of a flow #11234 by extracting the text field from Message/Data inputs.
• Addresses Run Flow component produces incorrect actions when tool_mode is enabled #11147 by exposing as tools only outputs that do not have outgoing edges.
• Adds regression tests for the listed bugs.

Summary by CodeRabbit

• Bug Fixes

  • Improved flow execution error handling with better logging and recovery mechanisms
  • Fixed output display to show only final results, skipping intermediate node outputs

• Improvements

  • Enhanced data extraction and input processing for flows
  • Better graph structure handling and caching behavior
  • Refined field formatting for improved readability

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Important

Review skipped

Auto incremental reviews are disabled on this repository.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

• 🔍 Trigger a full review

Walkthrough

This pull request refactors internal method naming and input handling in the RunFlowBaseComponent. Methods extracting tweaks are renamed to handle "ioputs", graph vertices are unfrozen after loading, error handling is added to cached flow execution, and output formatting is adjusted to skip non-terminal vertices.

Changes

Cohort / File(s)	Summary
Test Updates src/backend/tests/unit/base/tools/test_run_flow.py, src/backend/tests/unit/components/logic/test_run_flow_component.py	Method signature renames reflected in tests: _extract_tweaks_from_keyed_values → _extract_ioputs_from_keyed_values and _build_inputs_from_tweaks → _build_inputs_from_ioputs. Graph mock initialization updated with vertices and successor_map parameters. New test coverage for Data object handling in ioputs and vertex successor relationship validation.
Core Refactoring src/lfx/src/lfx/base/tools/run_flow.py	Internal method names updated from tweaks-oriented to ioputs-oriented terminology. Graph vertices explicitly unfrozen after loading. String formatting improved with spacing before parenthetical suffixes. List mutation optimized via explicit extend(). New helper methods _build_inputs() and _build_flow_tweak_data() introduced for input synthesis. _get_cached_run_outputs() and _run_flow_with_cached_graph() signatures simplified by removing tweaks/inputs parameters; internal construction now handled via helpers. Error handling added around flow execution with logging. Output formatting adjusted to skip vertices with outgoing edges. Tool-mode data extraction default changed to False.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

---

Important

Pre-merge checks failed

Please resolve all errors before merging. Addressing warnings is optional.

❌ Failed checks (1 error, 2 warnings, 1 inconclusive)

Check name	Status	Explanation	Resolution
Test Coverage For New Implementations	❌ Error	PR has incomplete test coverage for new _build_inputs method, unresolved issues in test_build_flow_tweak_data_merges_tool_tweaks, and missing tests for significant signature changes.	Add dedicated unit test for _build_inputs method, fix trailing whitespace and add mutation validation test for _build_flow_tweak_data, add comprehensive tests for signature changes to _get_cached_run_outputs and _run_flow_with_cached_graph.
Docstring Coverage	⚠️ Warning	Docstring coverage is 75.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Test Quality And Coverage	⚠️ Warning	The pull request has test quality and coverage issues including trailing whitespace, missing mutation verification tests, insufficient test coverage for helper methods, and lack of edge case testing.	Fix trailing whitespace, add mutation verification tests, create direct tests for _build_inputs, strengthen existing tests with actual data structure validation, and add edge case and error handling tests.
Title check	❓ Inconclusive	The title is vague and contains a spelling error ('progation' instead of 'propagation'), making it unclear what specific aspect of runflow toolmode input handling is being fixed.	Correct the spelling and be more specific about what the fix addresses, such as 'Fix: runflow toolmode input propagation in flow execution' or clarify which specific issue is being resolved.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Test File Naming And Structure	✅ Passed	Test files follow pytest naming conventions with 42 and 25 descriptive test functions organized in logical classes, comprehensive edge case coverage, proper fixtures, and async support.
Excessive Mock Usage Warning	✅ Passed	Test files demonstrate appropriate mock usage with 2.0-2.7 mocks per method, well below concerning thresholds. Helper method tests use minimal mocks while testing actual behavior; external dependencies appropriately mocked.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch fix/runflow-toolmode

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 2

🤖 Fix all issues with AI agents

In `@src/backend/tests/unit/base/tools/test_run_flow.py`:
- Around line 667-687: Remove the trailing whitespace in the test where
component._attributes is defined and add an assertion that _attributes is not
mutated by _build_flow_tweak_data: capture original_attributes =
dict(component._attributes), call component._build_flow_tweak_data(), then
assert component._attributes == original_attributes; if the implementation of
RunFlowBaseComponent._build_flow_tweak_data mutates _attributes, modify that
method to return merged tweak data without altering the original _attributes.

In `@src/lfx/src/lfx/base/tools/run_flow.py`:
- Around line 689-701: The method _build_flow_tweak_data mutates
self._attributes by assigning combined_values = self._attributes and calling
combined_values.update(tool_tweaks); instead, make combined_values a shallow
copy (e.g., copy dict(self._attributes) or use self._attributes.copy()) before
merging tool_tweaks so the original self._attributes is not modified, then
continue to call _extract_ioputs_from_keyed_values(combined_values); ensure you
also handle tool_tweaks.model_dump() as before and only update the copied
combined_values when tool_tweaks is a dict.

🧹 Nitpick comments (1)

src/lfx/src/lfx/base/tools/run_flow.py (1)

495-523: Graph mutation affects cached version; catch overly broad exceptions.

The code caches the graph in get_graph() and later passes it to process_tweaks_on_graph(), which mutates the graph in-place (iterates vertices and calls apply_tweaks_on_vertex() without copying). This means the cached version is modified, affecting subsequent executions. Consider adding a deepcopy before mutations or after retrieving from cache.

The try/except block catches all exceptions with # noqa: BLE001, potentially masking programming errors. get_graph() and run_flow() explicitly raise ValueError, but catching all exceptions and silently returning None can hide unexpected issues during debugging. Consider catching specific exception types.

[github-actions]

Frontend Unit Test Coverage Report

Coverage Summary

Lines	Statements	Branches	Functions
	17.54% (5050/28789)	10.96% (2432/22176)	11.63% (733/6299)

Unit Test Results

Tests	Skipped	Failures	Errors	Time
2036	0 💤	0 ❌	0 🔥	26.048s ⏱️

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 34.86%. Comparing base (47f4e80) to head (13a3880).
⚠️ Report is 1 commits behind head on main.

❌ Your project check has failed because the head coverage (41.60%) is below the target coverage (60.00%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #11348      +/-   ##
==========================================
+ Coverage   34.84%   34.86%   +0.02%     
==========================================
  Files        1420     1420              
  Lines       68200    68162      -38     
  Branches     9981     9974       -7     
==========================================
+ Hits        23762    23765       +3     
+ Misses      43213    43173      -40     
+ Partials     1225     1224       -1     

Flag	Coverage Δ
backend	54.21% <ø> (+0.01%)	⬆️
frontend	16.05% <ø> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/lfx/src/lfx/base/tools/run_flow.py	0.00% <ø> (ø)

... and 8 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[HzaRashid]

just note that im making a PR that not only fixes the bug but adds several integration tests to verify that the graph runtime behavior correctly reflects prior processing, for all processing helpers