packages/ai:
- fix: handle "sensitive" stop reason from Anthropic API
- fix: normalize tool call IDs with special characters for Responses API
- fix: add overflow detection for Bedrock, MiniMax, Kimi providers
- fix: 429 status is rate limiting, not context overflow
packages/tui:
- fix: refactored autocomplete state tracking
- fix: file autocomplete should not trigger on empty text
- fix: configurable autocomplete max visible items
- fix: improved table column width calculation with word-aware wrapping
packages/coding-agent:
- fix: preserve external config.yml edits on save (#1046 by @nicobailonMD)
- fix: resolve macOS NFD and curly quote variants in file paths
- Added `RequestOptions` interface to worker pool with support for `timeoutMs`, `signal`, and `transfer` properties.
- Added `stuckGracePeriodMs` configuration option to `WorkerPoolOptions` for graceful handling of stuck workers after timeout.
- Enhanced worker pool request method to accept `RequestOptions` object parameter instead of simple transfer array, enabling timeout and abort signal handling.
- Implemented comprehensive timeout and abort signal handling in worker pool with support for combined signals and grace period before force-terminating stuck workers.
- Removed html2text tool from coding agent in favor of native WASM html-to-markdown converter.
- Exported `RequestOptions` type from natives package public API.
- Added HTML to Markdown conversion functionality via new WASM binding in pi-natives crate with configurable options for content cleaning and image skipping.
- Exported htmlToMarkdown async function from @oh-my-pi/pi-natives package with worker pool support for efficient HTML to Markdown conversion.
- Integrated native WASM HTML to Markdown converter as the primary conversion method in fetch tool, replacing text-only rendering with markdown output.
- Added html-to-markdown-rs dependency to pi-natives crate for server-side HTML to Markdown conversion with preprocessing support.
- Added comprehensive test coverage for htmlToMarkdown function including headings, links, lists, code blocks, and conversion options.
- Added error field to LSP server status types throughout the mode context chain.
- Updated /session command to display the actual error message when an LSP server fails to start.
- Added logger.warn call during LSP warmup to log server startup failures to ~/.omp/logs/.
- Added Codex web search provider with OAuth credential management and SSE streaming support.
- Added Gemini web search provider with Google Search grounding and citation extraction.
- Extended WebSearchProvider type to include 'gemini' and 'codex' providers.
- Consolidated signal and timeout handling by replacing custom createRequestSignal utility with ptree.combineSignals across all scrapers and tools.
- Refactored ensureTool function to accept signal parameter through options object, enabling cancellation of tool installation and execution.
- Migrated all tool execution calls from Bun's $ template syntax to ptree.exec with explicit signal and error handling options.
- Simplified HTML rendering strategy in fetch.ts to use ptree.combineSignals for timeout management and added jina.ai integration as primary rendering method.
- Removed redundant abort checks throughout youtube.ts scraper, consolidating signal handling to rely on ptree.exec signal-based timeout mechanism.
- Added error handling for database initialization failures with detailed diagnostics including directory existence and writability checks.
- Improved directory creation in ensureDir() with proper error handling for EEXIST conditions and post-creation verification.
- Set restrictive file permissions (0o700) on created agent storage directories for improved security.
- Optimized applyLineResets() in TUI to use in-place array modification instead of creating new array via map().
- Added interactive `/debug` command with menu-driven interface for bug report triage and system diagnostics.
- Implemented CPU profiling workflow allowing users to capture performance profiles during issue reproduction.
- Added heap snapshot generation for memory issue analysis and debugging.
- Implemented debug report bundle creation with session data, logs, system info, and optional profiling artifacts.
- Added system information collection and display including OS, CPU, memory, and version details.
- Implemented artifact cache management with size statistics and cleanup functionality.
- Fixed incorrect byte and line limit calculations in file read tool that were applying user limits to total selected content instead of just user-limited content.
- Updated test to use SettingsManager.getGlobalShellConfig() instead of shell.getShellConfig() for consistent shell configuration access.
- Reorganized import statements across multiple files for consistency and alphabetical ordering.
- Changed firstLinePreviewChunks from let to const in read tool since it is only mutated via array methods.
- Refactored file reading logic to process bytes directly instead of decoding entire chunks into strings, improving memory efficiency and enabling better line boundary detection.
- Restructured line collection state management with dedicated helper functions (setupLineState, appendSegment, finalizeLine, decodeLine) for clearer control flow and byte tracking.
- Enhanced byte limit tracking to accurately count selected lines and bytes separately, fixing calculation of truncation metadata.
- Added firstLinePreview and firstLineByteLength fields to streamLinesFromFile return type to support better handling of oversized first lines.
- Improved firstLineExceedsLimit detection to use actual byte length instead of relying on empty collection state.
- Removed TUI layout engine exports from public API surface.
- Replaced Bun.file().stream() with fs/promises for manual buffer reading in file streaming.
- Added AbortSignal support to streamLinesFromFile for cancellation capability.
- Removed stdin monitoring feature that detected stalled input after 10 seconds.
- Removed cli-highlight and highlight.js dependencies from lock file.
- Rebuilt WASM module with reduced binary size (2.1% reduction).
- Extracted process management utilities from coding-agent shell module into dedicated procmgr module in utils package.
- Migrated shell configuration retrieval to SettingsManager class with new getGlobalShellConfig() static method.
- Replaced custom killProcessTree() implementation with new terminate() function supporting graceful shutdown with timeout and AbortSignal.
- Updated ptree.ChildProcess to use generic type parameter for input stream masking and delegated process termination to procmgr.terminate().
- Centralized process management across coding-agent modules to use SettingsManager and procmgr utilities instead of standalone shell functions.
- Added native syntax highlighting module powered by syntect library with ANSI color output support.
- Implemented highlightCode() function that applies syntax highlighting to code with language detection and customizable color themes.
- Implemented supportsLanguage() function to check if a given programming language is supported for syntax highlighting.
- Implemented getSupportedLanguages() function that returns list of all supported language identifiers.
- Replaced cli-highlight dependency with native syntect-based highlighting in coding-agent theme module.
- Added comprehensive language alias mappings supporting 50+ programming languages including TypeScript, Python, Rust, Go, C/C++, and web languages.
Multi-line bash commands (e.g., for loops with newlines) were being passed
as single "lines" to renderOutputBlock. Since visibleWidth() treats newlines
as width 0, truncateToWidth() calculated incorrect truncation points. The
result was rendered lines with embedded newlines that exceeded terminal width,
causing "Rendered line N exceeds terminal width" crashes.
Split section lines by newlines before processing so each line gets properly
truncated and wrapped with box borders.
- Fixed LSP servers showing as 'unknown' in status display when server warmup fails by using index-based lookup to correctly extract server name and file types.
- Fixed Read tool loading entire file into memory when offset/limit was specified by implementing streaming line-by-line processing that only collects requested range.
- Fixed slash command autocomplete applying stale completions when typing quickly by adding staleness check before applying completion on Enter/Return.
- Reorganized npm scripts to follow consistent naming pattern with separate TypeScript and Rust variants for check, lint, fmt, and fix operations.
- Updated release script to properly handle Rust workspace version updates in Cargo.toml files and regenerate lockfiles.
The nullish coalescing operator (??) was treating null as a fallback
trigger, causing getAskTimeout()'s null return value (meaning 'disabled')
to be replaced with DEFAULT_ASK_TIMEOUT_MS (30000ms).
Now explicitly checks for undefined to distinguish between:
- undefined: settingsManager unavailable → use default
- null: user explicitly disabled timeout → preserve null
- number: user-set timeout → use that value
- Added native text utilities in pi-natives (visible width, truncation, slicing, segment extraction)
- Wired pi-tui to use native implementations with short-string JS fast path
- Optimized key matching via cached parsed key ids
- Refactored visual truncation to reuse Text instances
- Reduced theme color lookups by switching from Map to plain records
- Regenerated WASM artifacts
Benchmarks (after vs before, lower is better):
- truncateToWidth/ansi: 20.37ms vs 38.13ms (2000 iters)
- sliceWithWidth/ansi: 5.12ms vs 20.39ms
- extractSegments/ansi: 8.92ms vs 27.60ms
- truncateToVisualLines: 3.51ms vs 411.55ms (500 iters)
- visibleWidth/ansi: 0.24ms vs 0.31ms
- Consolidated grep, find, and image processing into a new `@oh-my-pi/pi-natives` WASM package.
- Eliminated subprocess overhead and simplified JS tool implementations.
- Updated CI to include Rust workspace, linting, formatting, and WASM build steps.
- Documented `Promise.withResolvers()` and `using` declarations for wasm-bindgen.
- Added calculatePromptTokens() export to compute prompt-specific token counts from usage data.
- Added estimateEntriesTokens() helper to calculate total tokens across a range of session entries.
- Enhanced prepareCompaction() to dynamically adjust keepRecentTokens based on prompt-to-estimated token ratio for better context preservation.
- Improved compact() summary generation logic to preserve previous summaries when no new messages exist and handle edge cases with explicit fallback.
- Simplified ptree API by removing spawnGroup and spawnAttached variants, consolidating into single spawn function.
- Replaced AsyncQueue class with simpler pushStream utility function for stream management.
- Refactored ChildProcess class to use AbortController instead of callback-based signal handling.
- Renamed captureText method to wait and execText function to exec for clearer API semantics.
- Removed mode parameter from exec function, eliminating distinction between spawn modes.
- Updated all call sites across coding-agent to use simplified ptree API.
- Added configurable ask timeout and notification settings to control ask tool behavior and user notifications.
- Added AskSettings interface with timeout (in seconds, default 30) and notification method properties to settings manager.
- Added ask timeout and notification configuration options to settings UI with values for timeout (off, 15, 30, 60, 120 seconds) and notification methods (auto, bell, osc99, osc9, off).
- Refactored process execution across multiple tools (exec, fetch, grep, read, youtube scraper) to use centralized ptree.execText() API instead of custom implementations.
- Refactored ChildProcess class in ptree to use public readonly properties and Promise.withResolvers() for cleaner exit handling.
- Migrated process spawning from `cspawn` to `spawnGroup` and `spawnAttached` APIs with explicit resource management using TypeScript's `using` statement.
- Refactored `ChildProcess` class to support process group management with new `isProcessGroup` getter and `[Symbol.dispose]()` method for automatic cleanup.
- Removed `cspawn` from public API exports in pi-utils, replacing it with `spawnGroup` and `spawnAttached` functions.
- Simplified `ChildProcess.kill()` method signature by removing signal parameter and eliminated `killAndWait()` method in favor of async disposable pattern.
- Converted `killChild` function to async implementation using Bun's `$` template and sleep-based polling for process termination.
- Added outline option to dialog and select UI components for rendering outlined lists in extension dialogs and ask tool.
- Replaced ripgrep-based file listing with fs.glob for project scans in system-prompt, find, and read tools.
- Added timeout tracking to ask tool multi-select flow to prevent hanging on selection operations.
- Created new OutlinedList component in hook-selector for rendering bordered list containers.
- Implemented globPaths utility function in pi-utils for glob-based file discovery with gitignore support.
- Refactored hook selector and input overlay management to use TUI overlay API instead of manual container manipulation.
- Extracted dialog overlay options into a reusable constant with consistent styling (bottom-center anchor, 80% width, 70% max height).
- Added overlay handle tracking to properly manage overlay lifecycle and prevent duplicate overlays.
- Refactored runRg function signature to accept an options object with signal and timeoutMs properties instead of a direct signal parameter.
- Enhanced timeout error handling in grep.ts to catch ptree.TimeoutError exceptions and throw descriptive ToolError messages.
- Added proper resource disposal calls in extension-ui-controller.ts to dispose of hook selector and hook input before clearing editor containers.
- Simplified timeout handling by moving from AbortSignal-based approach to direct timeoutMs parameter in options object across system-prompt.ts, find.ts, and read.ts.
- Restructured project tree scanning to separate ripgrep-based and fallback implementations.
- Removed XML escaping function and changed tree output format from XML to plain text with visual indicators.
- Improved directory traversal to build complete directory contents map before rendering.
- Simplified tree rendering logic by removing XML tag generation and using text-based formatting.
- Removed incorrect millisecond-to-second auto-conversion logic in timeout handling across cursor, fetch, gemini-image, and ssh tools.
- Fixed timeout parameter interpretation to respect user-provided values without automatic unit conversion heuristics.
- Added gitignore-aware project tree scanning using ripgrep to respect .gitignore rules when discovering project structure.
- Implemented getGitignoreAllowedPaths() function to retrieve allowed files and directories based on gitignore rules with 5-second timeout.
- Enhanced scanProjectTree() to filter entries against gitignore-allowed paths while gracefully falling back to static filtering if ripgrep is unavailable.
- Restructured system prompt templates for improved clarity and consistency.
- Simplified XML tag structure in custom and system prompts by removing redundant wrapper elements and consolidating metadata.
- Updated skill and rule references to use inline path notation (e.g., `skill://<name>`, `rule://<name>`) instead of separate location tags.
- Improved prompt readability by reorganizing sections with clearer headers and more concise instructional text.
- Added project tree visualization to system prompts with configurable depth and entry limits.
- Implemented scanProjectTree function to recursively scan directory structure with intelligent truncation and sorting by modification time.
- Added renderProjectTree function to format scanned directory structure as XML with path collapsing for single-child directories.
- Integrated projectTree snapshot into both custom and main system prompt templates with conditional rendering.
- Added XML escaping utility to safely render file paths and names in prompt context.
- Added reset() method to CountdownTimer class to allow resetting countdown to initial value.
- Added initialMs field to CountdownTimer to track original timeout duration.
- Integrated countdown reset into HookSelectorComponent to reset timer on user interaction.
- Added SkillMessageComponent to render /skill command messages as compact skill entries with expandable/collapsible UI showing skill metadata and optional full prompt content.
- Added promptCustomMessage() public method to AgentSession to support custom message types with metadata and details properties.
- Added SKILL_PROMPT_MESSAGE_TYPE constant and SkillPromptDetails interface to support skill-specific message handling.
- Enhanced patch tool documentation with guidance on structured block editing, error recovery, and best practices for maintaining code integrity.
- Refactored message handling in AgentSession by extracting _promptWithMessage() private method to consolidate common logic between prompt() and promptCustomMessage().
- Added model preference matching system with ModelMatchPreferences and ModelPreferenceContext interfaces to enable intelligent model selection based on usage history and provider preferences.
- Implemented buildPreferenceContext and pickPreferredModel functions to rank and select models based on usage order, provider history, and deprioritization settings.
- Enhanced parseModelPattern function to accept optional ModelMatchPreferences parameter, enabling model resolution to consider user preferences and usage history.
- Updated model matching logic to prioritize models based on historical usage patterns and provider preferences when multiple candidates match a pattern.
- Converted wrapper class properties to dynamic getters that delegate to underlying tool objects.
- Added optional chaining operators for safer null/undefined access in JSON schema property handling.
- Changed output metadata wrapper to use Object.create() with property descriptors instead of object spread syntax.
- Added model-specific edit variant configuration allowing users to specify 'patch' or 'replace' mode per model pattern.
- Added `getEditVariantForModel()` method to determine appropriate edit mode based on user settings and default patterns.
- Made EditTool's `patchMode`, `description`, and `parameters` dynamic properties that adapt based on the active model.
- Added fallback mechanism in model resolution to support tool creation before agent initialization.
- Added ModelStatsTable component to display model statistics with columns for model, provider, requests, cost, tokens, throughput, and error rate.
- Added 'models' tab to stats dashboard for viewing aggregated model performance metrics.
- Added TTFT (Time to First Token) metric display in request details and overview statistics.
- Added Tokens/s (throughput) metric calculation and display in request list and model statistics.
- Improved code formatting and accessibility by adding semantic HTML attributes (type='button', role, aria-modal) and reformatting style objects for readability.
- Refactored loadData function to use useCallback hook for better performance optimization.
- Added artifact allocation for truncated fetch responses to preserve full content in output artifacts.
- Integrated truncateHead utility to apply consistent truncation limits (DEFAULT_MAX_BYTES and DEFAULT_MAX_LINES) to fetch tool output.
- Refactored output formatting into buildOutput helper function to support both truncated and full content rendering.
- Added Kimi Code provider integration with OAuth device authorization flow and token management.
- Added four new Kimi Code models (kimi-for-coding, kimi-k2, kimi-k2-turbo-preview, kimi-k2.5) with reasoning support and 262K context window.
- Added kimiUsageProvider for fetching and caching Kimi Code API usage quota information.
- Added kimi-code login command to CLI for OAuth authentication with Kimi Code.
- Updated openai-completions provider to support Kimi-specific headers and cached token formats.
- Updated MiniMax-M2 model pricing: input 1.2->0.6, output 1.2->3, cacheRead 0.6->0.1.
- Migrated import paths from absolute package imports to relative imports throughout coding-agent modules.
- Consolidated related imports in tools/index.ts to reduce import statement count.
- Removed guideline about avoiding relative parent imports from AGENTS.md documentation.
- Reformatted code examples in AGENTS.md for improved readability with multi-line formatting.
- Added timeout support to hook selector with auto-selection behavior on timeout.
- Extended ExtensionUIDialogOptions to include timeout and tui configuration properties.
- Changed hook selector timeout behavior from cancellation to auto-selecting the currently highlighted option.
- Updated UIContext.select() method signature to accept timeout option for automated selection scenarios.
- Renamed 'output' parameter to 'schema' throughout task configuration and documentation for clarity.
- Updated task schema definition to use 'schema' field with improved description clarifying it defines expected response structure.
- Restructured task summary output format to consolidate metadata into tag attributes and simplify XML structure.
- Added truncation detection for task output exceeding 5000 character threshold with truncated flag.
- Removed charCount from task metadata, retaining only lineCount and charSize properties.
- Extracted tool choice mapping logic into dedicated utility module for provider-specific format conversions.
- Refactored task summary generation to use template-based rendering instead of string concatenation.
- Migrated system prompt generation to use Handlebars templates with structured data instead of hardcoded strings.
- Extracted JTD to TypeScript conversion logic into dedicated utility module for schema documentation.
- Consolidated prompt template rendering across multiple modules using unified renderPromptTemplate function.
- Simplified type assertions in stream module to use generic OptionsForApi type instead of provider-specific types.
- Renamed 'complete' tool to 'submit_result' throughout codebase for clarity and consistency.
- Updated all tool references, function names, and variable names from 'complete' to 'submit_result' in executor, render, and tools modules.
- Renamed CompleteTool class to SubmitResultTool and CompleteDetails interface to SubmitResultDetails.
- Updated agent prompts and documentation to reference the new 'submit_result' tool name.
- Reorganized task.md prompt to move critical guidance to the top and restructure instructions for better clarity.
- Added available tools list to system prompt for agent coherence.
- Improved output preview logic to show full output for smaller results and truncate larger outputs to 10 lines.
- Added session compaction auto-continue feature that automatically sends a synthetic continuation prompt after compaction completes when enabled in settings.
- Added tool output pruning system that intelligently removes verbose tool outputs from session history during compaction to reduce token usage while preserving critical information.
- Added short summary generation for compaction results, providing concise PR-style summaries of conversation changes alongside full summaries.
- Added session.compacting extension hook that allows extensions to customize compaction prompts, add context, and preserve metadata during summarization.
- Added synthetic message support to mark system-injected messages (like auto-continue prompts) and prevent them from being persisted to editor history.
can1357