True root cause: When reopenParentFromDelegation created a NEW user message
with only the new_task tool_result, validateAndFixToolResultIds would see
the assistant message expecting BOTH tool_results (e.g., update_todo_list + new_task)
but the new message only had new_task, causing a placeholder to be generated
for update_todo_list.
Fix: Find the existing user message that was created by flushPendingToolResultsToHistory
(which already has other tool_results like update_todo_list) and APPEND the
new_task tool_result to it instead of creating a new message.
This ensures all tool_results for a turn are in the same user message,
matching what the Anthropic API expects.
Root cause: In NewTaskTool.execute(), pushToolResult was called AFTER
delegateParentAndOpenChild(), but delegateParentAndOpenChild disposes
the parent task. This meant the tool_result was lost, and
flushPendingToolResultsToHistory generated a placeholder "interrupted"
tool_result instead.
Fix: Push the tool_result BEFORE calling delegateParentAndOpenChild.
Since the child taskId is not yet known, use a generic message.
The actual completion result is injected by reopenParentFromDelegation
when the child completes.
Also keep the defensive duplicate check in reopenParentFromDelegation
for backward compatibility with any existing task histories.
- Check ALL user messages in parentApiMessages for existing tool_result
with the same tool_use_id before appending a new one
- Previously only checked the last message, missing duplicates in earlier messages
- Log warning when skipping duplicate tool_result for debugging
- Add tests for duplicate detection and normal case scenarios
* feat(condense): add smart code folding with tree-sitter signatures
At context condensation time, use tree-sitter to generate folded code
signatures (function definitions, class declarations) for files read
during the conversation. Each file is included as its own <system-reminder>
block in the condensed summary, preserving structural awareness without
consuming excessive tokens.
- Add getFilesReadByRoo() method to FileContextTracker
- Create generateFoldedFileContext() using tree-sitter parsing
- Update summarizeConversation() to accept array of file sections
- Each file gets its own content block in the summary message
- Add comprehensive test coverage (12 tests)
* fix: skip tree-sitter error strings in folded file context
- Add isTreeSitterErrorString helper to detect error messages
- Skip files that return error strings instead of embedding them
- Add test for error string handling
* refactor: move generateFoldedFileContext() inside summarizeConversation()
- Update summarizeConversation() to accept filesReadByRoo, cwd, rooIgnoreController instead of pre-generated sections
- Move folded file context generation inside summarizeConversation() (lines 319-339)
- Update ContextManagementOptions type and manageContext() to pass new parameters
- Remove generateFoldedFileContext from Task.ts imports - folding now handled internally
- Update all tests to use new parameter signature
- Reduces Task.ts complexity by moving folding logic to summarization module
* fix: prioritize most recently read files in folded context
Files are now sorted by roo_read_date descending before folded context
generation, so if the character budget runs out, the most relevant
(recently read) files are included and older files are skipped.
* refactor: improve code quality in condense module
- Convert summarizeConversation to use options object instead of 11 positional params
- Extract duplicated getFilesReadByRoo error handling into helper method
- Remove unnecessary re-export of generateFoldedFileContext
- Update all test files to use new options object pattern
* fix: address roomote feedback - batch error logging and early budget exit
---------
Co-authored-by: Roo Code <roomote@roocode.com>
Co-authored-by: daniel-lxs <ricciodaniel98@gmail.com>
fix: prevent false validation error for local Ollama models
The validation logic was checking against an empty router models object
that was initialized but never populated for Ollama. This caused false
validation errors even when models existed locally.
Now only validates against router models if they actually contain data,
preventing the false error when using local Ollama models.
Fixes ROO-581
Co-authored-by: Roo Code <roomote@roocode.com>
- Modified VsCodeLmHandler.internalCountTokens() to create temporary cancellation tokens when needed
- Token counting now works both during and outside of active requests
- Added 4 new tests to verify the fix and prevent regression
- Resolves issue where VS Code LM API users experienced context overflow errors
* Taskheader
* Subtask messages
* View subtask
* subtasks in history items
* i18n
* Table
* Lighter visuals
* bug
* fix: Align tests with implementation behavior
* refactor: extract CircularProgress component from TaskHeader
- Created reusable CircularProgress component for displaying percentage as a ring
- Moved inline SVG calculation from TaskHeader.tsx to dedicated component
- Added comprehensive tests for CircularProgress component (14 tests)
- Component supports customizable size, strokeWidth, and className
- Includes proper accessibility attributes (progressbar role, aria-valuenow)
* chore: update StandardTooltip default delay to 600ms
As mentioned in the PR description, increased the tooltip delay to 600ms
for less intrusive tooltips. The delay is still configurable via the
delay prop for components that need a different value.
---------
Co-authored-by: Roo Code <roomote@roocode.com>
- Remove getToolDescription() method from MultiSearchReplaceDiffStrategy
- Remove getToolDescription() from DiffStrategy interface
- Remove unused ToolDescription type from shared/tools.ts
- Remove unused eslint-disable directive
- Update test mocks to remove getToolDescription references
- Remove getToolDescription tests from multi-search-replace.spec.ts
Native tools are now defined in src/core/prompts/tools/native-tools/ using
the OpenAI function format. The removed code was dead since XML-style tool
calling was replaced with native tool calling.
* fix: convert orphaned tool_results to text blocks after condensing
When condensing occurs after assistant sends tool_uses but before user responds,
the tool_use blocks get condensed away. User messages containing tool_results that
reference condensed tool_use_ids become orphaned and get filtered out by
getEffectiveApiHistory, causing user feedback to be lost.
This fix enhances the existing check in addToApiConversationHistory to detect when
the previous effective message is not an assistant and converts any tool_result
blocks to text blocks, preventing them from being filtered as orphans.
The conversion happens at the latest possible moment (message insertion) because:
- Tool results are created before we know if condensing will occur
- We need actual effective history state to make the decision
- This is the last checkpoint before orphan filtering happens
* Only include environment details in summary for automatic condensing
For automatic condensing (during attemptApiRequest), environment details
are included in the summary because the API request is already in progress
and the next user message won't have fresh environment details injected.
For manual condensing (via condenseContext button), environment details
are NOT included because fresh details will be injected on the very next
turn via getEnvironmentDetails() in recursivelyMakeClineRequests().
This uses the existing isAutomaticTrigger flag to differentiate behavior.
---------
Co-authored-by: Hannes Rudolph <hrudolph@gmail.com>
* chore: remove MULTI_FILE_APPLY_DIFF experiment
Remove the 'Enable concurrent file edits' experimental feature that
allowed editing multiple files in a single apply_diff call.
- Remove multiFileApplyDiff from experiment types and config
- Delete MultiFileSearchReplaceDiffStrategy class and tests
- Delete MultiApplyDiffTool wrapper and tests
- Remove experiment-specific code paths in Task.ts, generateSystemPrompt.ts, and presentAssistantMessage.ts
- Remove special handling in ExperimentalSettings.tsx
- Remove translations from all 18 locale files
The existing MultiSearchReplaceDiffStrategy continues to handle
multiple SEARCH/REPLACE blocks within a single file.
* fix: remove unused EXPERIMENT_IDS/experiments import from Task.ts
Addresses review feedback: removes the unused imports from
src/core/task/Task.ts that were left over after removing the
MULTI_FILE_APPLY_DIFF experiment routing code.
- Remove powerSteering from experimentIds array and schema in packages/types
- Remove POWER_STEERING from EXPERIMENT_IDS and experimentConfigsMap
- Remove power steering conditional block from getEnvironmentDetails
- Remove POWER_STEERING entry from all 18 locale settings.json files
- Update related test files to remove power steering references
