Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add truncateToolOutput helper in truncation.ts to centralize threshold reading, file saving, and telemetry logging
- Refactor shell.ts to use the new helper, removing duplicate code
- Add truncation support for MCP tool output while preserving non-text content (images, audio, resources)
- Refactor getDisplayFromParts to work on transformed Part[] instead of raw MCP response
This reduces code duplication and ensures consistent truncation behavior across shell and MCP tools.
- Added candidatesTokens prop to LoadingIndicator for displaying token counts.
- Updated formatting to show elapsed time and token counts inline.
- Refactored tests to validate new token display functionality and formatting changes.
- Introduced formatTokenCount utility for consistent token count representation.
This improves user feedback during loading states by providing clearer information on token usage.
- Add hasExplicitOutputLimit() to detect models with defined output limits
- For known models: cap user max_tokens to model limit (avoid API errors)
- For unknown models (deployment aliases, self-hosted): respect user config
- Update tests to cover new behavior
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add os.tmpdir() to allowed paths in read-file tool
- Add tests for reading files from OS temp directory
- Add terminal capture scenario for PR review testing
This supports the PR review workflow which saves context to temp files.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Increase DEFAULT_OUTPUT_TOKEN_LIMIT from 16K to 32K
- Remove auto-detection from modelsConfig, apply at provider level
- Use conservative default (min of model limit and 32K) when user hasn't configured max_tokens
- Respect user configuration but cap at model's max output limit to avoid API errors
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Checkout PR branch instead of remote viewing for full file access
- Save PR context to temp file to avoid repeating in agent prompts
- Add guidance to prevent 4x diff duplication across agents
- Include environment restoration step after review
This enables agents to read files directly and use git diff against base branch,
improving review quality and reducing prompt bloat.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Removed client-side filtering for search queries; fuzzy search is now handled by the backend.
- Enhanced file search initialization and caching mechanisms in FileMessageHandler.
- Added file watchers for cache invalidation on file system changes.
- Updated completion trigger logic to prioritize '@' over '/' for path-like queries.
- Reset last query on file selection to ensure fresh search results.
This refactor improves search efficiency and maintains accurate file references in the application.
Task tools spawn independent sub-agents with no shared mutable state,
making them safe to run in parallel. This change executes all task
tools concurrently while keeping other tools sequential to preserve
any implicit ordering the model may rely on.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Remove unused outputFile property from AgentToolResultEvent and its
associated test case. This property is not needed for agent tool
result handling.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add helper functions for better code organization (createPrintableKey,
getCompleteCsiSequenceLength, parsePlainTextPrefix)
- Drop unsupported Kitty CSI-u keys without blocking subsequent input
- Recover plain text that arrives in same chunk after unsupported CSI-u keys
- Add comprehensive tests for edge cases (CAPS_LOCK, event metadata variants)
Improves robustness of Kitty keyboard protocol parsing by gracefully
handling unsupported key codes and ensuring plain text input is not lost.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Add needsUtf8Bom() to detect when UTF-8 BOM is needed based on file
extension and system code page
- PowerShell 5.1 on non-UTF-8 Windows systems (e.g. GBK) requires BOM
to read scripts correctly
- Remove default UTF8 encoding; undefined now triggers auto-detection
- Add tests for needsUtf8Bom() covering Windows/non-Windows scenarios
This ensures PowerShell scripts are written with UTF-8 BOM on systems
that need it, fixing character encoding issues for non-ASCII content.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Refactor the vi.mock for 'os' to use a simpler direct mock object
instead of the importOriginal pattern, making the test setup more concise.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Make CRLF conversion for .bat/.cmd files Windows-only
- Extract PowerShell UTF-8 prefix into reusable function
- Replace custom UTF-8 validation with Node.js built-in isUtf8()
This ensures .bat/.cmd files are only converted on Windows where cmd.exe
actually requires CRLF, and reduces code duplication for shell encoding.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Update JSDoc to make clear that detectEncodingFromBuffer only performs
chardet statistical detection and returns null on failure. Callers like
getCachedEncodingForBuffer are responsible for providing fallback logic.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Move system encoding fallback from detectEncodingFromBuffer into
getCachedEncodingForBuffer for clearer responsibility. Remove unused
WINDOWS_UTF8_CODE_PAGE export and inline the value.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Prefix PowerShell commands with [Console]::OutputEncoding=UTF8
- Re-detect encoding on full buffer after streaming completes
- Move system encoding fallback into detectEncodingFromBuffer
This ensures CJK and other non-ASCII characters are correctly decoded
on Windows systems with non-UTF-8 system codepages (e.g., GBK/CP936).
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
- Fix generationConfigSources to preserve existing source info when auto-detecting max_tokens
- Add unit tests for max_tokens fallback logic
When modelProviders config does not specify samplingParams.max_tokens,
requests to non-Qwen models (Claude, GPT, Gemini, etc.) omit max_tokens
entirely. Many APIs default to a small value (e.g., Anthropic via
VertexAI defaults to 4096), causing long responses to be truncated
mid-generation — often breaking tool call parameters.
Fix: apply tokenLimit(model, 'output') as a fallback in
applyResolvedModelDefaults(), following the same pattern already used
for contextWindowSize and modalities auto-detection.
Output limits from tokenLimits.ts:
- Claude Opus 4.6: 128K
- Claude Sonnet 4.6 / fallback: 64K
- GPT-5.x: 128K
- Gemini 3.x: 64K
- Qwen 3.5: 64K
Made-with: Cursor
- Rename outputEncoding to detectedEncoding for clarity
- Add isValidUtf8 helper using TextDecoder in fatal mode
- Restructure detection: UTF-8 → chardet → system encoding
- Update tests to use non-UTF-8 bytes for accurate testing
This prevents misclassifying UTF-8 output as legacy codepages on systems
where the system encoding (e.g., GBK) could also decode those bytes.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Remove wrapCommandForWindowsEncoding and forceUtf8Output parameter,
relying solely on getCachedEncodingForBuffer for encoding detection.
This simplifies the shell execution flow by removing the chcp 65001
command prefixing approach.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
