Commit Graph

68 Commits

  • Add --mode flag for CLI output control (text/json/rpc)
    - text mode: only outputs final assistant message text (default)
    - json mode: streams all events as JSON (same as session manager writes)
    - rpc mode: JSON output + listens for JSON input on stdin for headless operation
    - Suppress informational messages in json/rpc modes
  • Improve edit tool diff display with context-aware rendering
    - Add generateDiffString() function in edit tool to create unified diffs with line numbers and 4 lines of context
    - Store only the formatted diff string in tool result details instead of full file contents
    - Update tool-execution renderer to parse and colorize the diff string
    - Filter out message_update events from session saving to prevent verbose session files
    - Add markdown nested list and table rendering tests
  • Add proper truncation notices and comprehensive tests for read tool
    **Improved output messages:**
    1. File fits within limits: Just outputs content (no notices)
    2. Lines get truncated: Shows "Some lines were truncated to 2000 characters for display"
    3. File doesn't fit limit: Shows "N more lines not shown. Use offset=X to continue reading"
    4. Offset beyond file: Shows "Error: Offset X is beyond end of file (N lines total)"
    5. Both truncations: Combines both notices with ". " separator
    
    **Comprehensive test coverage:**
    - Files within limits (no notices)
    - Large files (line truncation)
    - Long lines (character truncation)
    - Offset parameter
    - Limit parameter
    - Offset + limit together
    - Invalid offset (out of bounds)
    - Combined truncations (both notices)
    
    All 17 tests passing ✓
  • Add line limits and truncation to read tool
    - Default limit of 2000 lines per read
    - Maximum line length of 2000 characters (truncates longer lines)
    - Add optional offset and limit parameters for pagination
    - Show truncation notice with instruction to continue reading
    - Plain text output (no line numbers)
    
    Examples:
    - `read path/to/file` - reads first 2000 lines
    - `read path/to/file offset=2001` - continues from line 2001
    - `read path/to/file limit=100` - reads only 100 lines
    
    This prevents extremely large files from overwhelming context.
  • Match export output style to thinking level display
    Changed export success/error messages to match the style used for thinking level changes:
    - Use Spacer(1) instead of empty Text for proper blank line spacing
    - Use chalk.dim() for success message instead of chalk.green()
    - Add left padding (1, 0) to match other system messages
    - Removed checkmark/x symbols for cleaner look
    
    Now "Session exported to: filename.html" appears dimmed with proper spacing, just like "Thinking level: off"
  • Make HTML export header minimal and stack info vertically
    **Header improvements:**
    - Changed title from "pi coding-agent session" to "pi v{version}"
    - Smaller, more compact header (16px instead of 20px)
    - Info items now stack vertically with consistent labels (80px width)
    - Reduced padding and spacing for cleaner look
    - Labels aligned to left, values aligned to baseline
    
    The header is now much less cramped and easier to read.
  • Add click-to-expand and system prompt display to HTML export
    **Click-to-expand functionality:**
    - Truncated tool outputs are now clickable to show full content
    - Works for bash (>5 lines), read (>10 lines), write (>10 lines)
    - Displays "... (N more lines) - click to expand" hint in cyan
    - Smooth toggle between preview and full output
    - Cursor changes to pointer on hover
    
    **System prompt and tools display:**
    - Added system prompt section with yellow header on dark brownish background
    - Shows all available tools with names and descriptions
    - Both sections appear before messages for context
    - Subtle yellow/brown theme (rgb(60, 55, 40)) that's not too bright
    
    The HTML export now provides better interactivity for large outputs and gives full context about the agent's configuration.
  • Rewrite HTML export to match TUI visual style exactly
    Complete overhaul of export-html.ts to replicate the TUI renderer's visual design:
    
    **Exact TUI color matching:**
    - User messages: rgb(52, 53, 65) dark slate background
    - Tool pending: rgb(40, 40, 50) dark blue-gray
    - Tool success: rgb(40, 50, 40) dark green
    - Tool error: rgb(60, 40, 40) dark red
    - Body background: rgb(24, 24, 30) very dark
    - Text colors matching chalk: cyan for paths, dimmed gray for output
    
    **Tool formatting (matching ToolExecutionComponent):**
    - bash: Bold `$ command`, dimmed output (max 5 lines)
    - write: Bold header with cyan path, line count, content preview (max 10 lines)
    - read: Bold header with cyan path, content preview (max 10 lines)
    - edit: Bold header with cyan path, unified diff with line numbers
    - All tools show "... (N more lines)" indicators
    
    **Additional improvements:**
    - Use session filename + .html for default output
    - Shorten paths with tilde notation (~)
    - Replace tabs with 3 spaces
    - Monospace font matching terminal aesthetic
    - Minimal, clean design with no borders/boxes
    - Print-friendly styles
    
    The HTML export now looks identical to the TUI version.
  • Add /export command to export sessions as self-contained HTML
    - Add exportSessionToHtml function that generates beautifully formatted HTML exports
    - HTML includes session metadata, all messages, tool calls, tool results, thinking blocks, and images
    - Support for ANSI color codes in tool output (converted to HTML)
    - Self-contained with inline CSS (dark theme, responsive design, print-friendly)
    - Add /export slash command to TUI with optional filename parameter
    - Add agent and coding-agent to dev script for watch mode
    - Increment coding-agent version to 0.6.1
    
    Usage: /export [optional-filename.html]
  • Handle FinishReason.NO_IMAGE and fix optional chaining
    - Add NO_IMAGE to error finish reasons in Google provider
    - Fix non-null assertion after optional chaining in Anthropic provider
    - Migrate biome config to 2.3.5
    - Ignore Tailwind CSS file from biome checks
    - Bump all packages to version 0.6.0
  • Fix coding agent tools to return error content instead of throwing
    Tools now resolve with error messages in content blocks rather than
    rejecting the promise. This matches the expected behavior where tools
    always return a result, with errors indicated in the text content.
    
    - read: Return error content for missing files
    - edit: Return error content for missing files, text not found, multiple matches
    - bash: Return error content for command failures, timeouts, and aborts
    - All tools now include required details field in error results
  • Add image support in tool results across all providers
    Tool results now use content blocks and can include both text and images.
    All providers (Anthropic, Google, OpenAI Completions, OpenAI Responses)
    correctly pass images from tool results to LLMs.
    
    - Update ToolResultMessage type to use content blocks
    - Add placeholder text for image-only tool results in Google/Anthropic
    - OpenAI providers send tool result + follow-up user message with images
    - Fix Anthropic JSON parsing for empty tool arguments
    - Add comprehensive tests for image-only and text+image tool results
    - Update README with tool result content blocks API
  • Search through all messages in session, not just first message
    - SessionManager now collects text from all user and assistant messages
    - Added allMessagesText field containing concatenated message text
    - Search now matches against all messages in a session
    - More powerful session discovery (e.g., search "write file" finds any session discussing file writing)
  • Add search, scrolling, spacing, and Ctrl+C exit to session selector
    - Add search input at top - filters sessions as you type
    - Add scrolling with max 5 visible sessions at a time
    - Add scroll indicator showing position (e.g., (3/15))
    - Add blank line spacing between session items
    - Ctrl+C now exits process (process.exit(0)) instead of canceling
    - Escape still cancels and returns to main menu
    - Enter in search box selects current session
  • Redesign session selector with multi-line layout
    - Custom SessionList component with 2 lines per item
    - First line: message text (bold when selected)
    - Second line: metadata (time · message count) in dim
    - Blue › cursor for selected item
    - Added "Resume Session" header
    - Handle Ctrl+C (‹x03›) in SelectList for consistency
    - Improved date formatting (38 minutes ago, 8 hours ago, etc.)
  • Fix session selector: handle Ctrl+C and viewport width
    - Add SIGINT handler to allow Ctrl+C to exit during session selection
    - Remove hardcoded 60-char truncation, let SelectList handle dynamic truncation
    - Simplify date format (5m, 2h, 3d instead of "5m ago")
    - Make metadata more compact (5msg instead of "5 msgs")
  • Add --resume flag with session selector
    - New SessionSelectorComponent to browse and select sessions
    - Lists sessions sorted by last modified date
    - Shows first message, created/modified dates, message count
    - Automatically truncates long messages and formats dates
    - Adds --resume/-r flag to CLI
    - Session selector integrates with main flow
  • Add custom session events for thinking level and model changes
    Session manager changes:
    - Add ThinkingLevelChangeEntry and ModelChangeEntry types
    - Add saveThinkingLevelChange() and saveModelChange() methods
    - Update loadThinkingLevel() to also check for thinking_level_change events
      (not just session headers)
    
    TUI changes:
    - Pass SessionManager to TuiRenderer constructor
    - Call saveThinkingLevelChange() when user changes thinking level via /thinking
    - Store session manager as instance variable for future use
    
    This ensures thinking level changes during a session are persisted
    and correctly restored on --continue.
  • Fix viewport width issues in thinking selector and text rendering
    Thinking selector fix:
    - Replace hardcoded 50-character borders with DynamicBorder component
      that adjusts to viewport width
    - Prevents crash when terminal is resized narrow
    
    Text/Markdown rendering fixes:
    - Fix Markdown wrapSingleLine to check if adding next character would
      exceed width BEFORE adding it (was checking AFTER, causing lines to
      be 1 character too long)
    - Add word truncation in Text component for words longer than
      contentWidth (prevents long unbreakable words from exceeding width)
    
    These fixes prevent "Rendered line exceeds terminal width" errors.
  • Fix SelectList crash on narrow terminal and thinking level restoration
    SelectList fixes:
    - Use hardcoded visual width for arrow prefix instead of string length
      which includes ANSI codes
    - Truncate scroll indicator text to prevent exceeding terminal width
    - Fixes crash when terminal is resized narrow and arrow keys are pressed
    
    Thinking level fix:
    - Restore thinking level even if it's "off" (removed !== "off" check)
    - This ensures the thinking level selector shows the correct state on continue
  • Show "Aborted" text when aborting text generation during streaming
    Update streaming component with final message in message_end event.
    The final message includes the stopReason, which allows the component
    to render "Aborted" text for aborted text generation.
    
    Now behavior is consistent:
    - Abort during text generation (no tool calls) → shows red "Aborted"
    - Abort during tool call streaming → tool components turn red
  • Fix aborted tool display when continuing sessions
    Two fixes for rendering aborted tools in --continue mode:
    
    1. In renderInitialMessages: Check if assistant message was aborted/errored
       and immediately mark tool execution components as failed instead of
       leaving them in pending state.
    
    2. In AssistantMessageComponent: Don't show "Aborted" text when there are
       tool calls in the message, since the tool execution components will
       show the error state themselves. This prevents duplicate error messages.
    
    Now aborted tools show properly as red with "Operation aborted" message,
    without the duplicate "Aborted" text above them.
  • Mark pending tool components as aborted when streaming is interrupted
    When user presses Esc during tool argument streaming, the assistant
    message ends with stopReason="aborted" but tool execution never starts.
    Now we check for aborted/error stopReasons in message_end and update
    all pending tool execution components to show red error state with
    "Operation aborted" message.
    
    This fixes the issue where tool components stayed stuck in blue/pending
    state after aborting during JSON streaming.
  • Make file operations properly abortable with async operations
    Replace synchronous file operations with async Promise-based operations
    that listen to abort signals during execution:
    - read, write, edit now use fs/promises async APIs
    - Add abort event listeners that reject immediately on abort
    - Check abort status before and after each async operation
    - Clean up event listeners properly
    
    This ensures pressing Esc during file operations shows red error state.
  • Add abort signal handling to read, write, and edit tools
    All tools now check the abort signal before executing and throw
    "Operation aborted" error if the signal is already aborted.
    This ensures consistent abort behavior across all tools.
  • Fix tab rendering in TUI components
    Replace tabs with 3 spaces for consistent rendering and width calculation:
    - Updated visibleWidth() to normalize tabs before measuring
    - Updated Text and Markdown components to replace tabs when rendering
    - Updated tool-execution display for read/write tools to replace tabs
    
    This fixes background color rendering issues when displaying files with tab indentation.
  • Add thinking level persistence and fix UI issues
    - Save and restore thinking level when continuing sessions
    - Fix thinking level confirmation message spacing and styling
    - Fix thinking text wrapping to preserve ANSI formatting across lines
  • Improve tool execution rendering and error handling
    - Show tool execution components immediately when tool calls appear in streaming
    - Update components with streaming arguments as they come in
    - Handle incomplete/partial arguments gracefully with optional chaining
    - Fix error handling: tools now throw exceptions instead of returning error messages
    - Fix bash abort handling to properly reject on abort/timeout
    - Clean up error display
  • Remove StreamingMessageComponent - just use AssistantMessageComponent
    - StreamingMessageComponent was just a wrapper around AssistantMessageComponent
    - AssistantMessageComponent now handles its own stats rendering
    - Made AssistantMessageComponent updatable with updateContent()
    - Removed duplicate stats handling code from tui-renderer
    - All stats are now managed by the component itself
  • Refactor TUI into proper components
    - Create UserMessageComponent - handles user messages with spacing
    - Create AssistantMessageComponent - handles complete assistant messages
    - Create ThinkingSelectorComponent - wraps selector with borders
    - Add setSelectedIndex to SelectList for preselecting current level
    - Simplify tui-renderer by using dedicated components
    - Much cleaner architecture - each message type is now a component
  • Simplify tool execution component structure with proper padding
    - Use single Text component with paddingY=1 for green background box
    - Add spacer line with transparent background at top
    - Header and content in same component with blank line separator
    - Move error markers to header line instead of after content
  • Fix tool execution formatting and colors
    - Add blank spacer line with no background before tool execution
    - Apply colored background to header (read/write/edit line)
    - Apply same colored background to content
    - Use lighter gray (chalk.dim) for file content
    - Show first 10 lines of content for write operations
    - Use cyan color for file paths (matching markdown code color)
  • Replace Markdown with Text component for tool execution
    - Add background color support to Text component
    - Replace Markdown component with Text in ToolExecutionComponent
    - Use chalk.bold for formatting instead of markdown syntax
    - Remove unnecessary markdown parsing overhead
  • Fix read/write tool output formatting
    - Remove code fences (```) from read output - show raw content
    - Show first 10 lines instead of 5 for read
    - Only show line count for write if > 10 lines
    - Remove indentation from read output
    - Cleaner, more minimal tool output display
  • Refactor TUI components into separate files
    - Move TUI components into src/tui/ folder
    - Split out CustomEditor, StreamingMessageComponent, ToolExecutionComponent, FooterComponent
    - Trim assistant message text content
    - Add newline after per-message token/cost stats
    - Improve code organization and maintainability
  • Clean up bash output formatting
    - Remove "Command aborted by user" message (Aborted already shown separately)
    - Remove STDOUT/STDERR labels - just show output cleanly
    - Stderr is now separated by newline without labels
    - Timeout and error messages moved to end of output
  • Fix bash abort to kill entire process tree immediately
    - Switch from exec() to spawn() with detached: true
    - Create new process group for spawned commands
    - Kill entire process group with process.kill(-pid) on abort
    - This ensures commands like "sleep 4 && echo hello" abort immediately
    - Previous implementation only killed parent shell, leaving subprocesses running
  • Add thinking trace support and improve bash output formatting
    - Render thinking traces in dark gray italic in final messages
    - Show thinking as italic markdown in streaming view
    - Remove code fences from bash output for cleaner minimal display
    - Remove superfluous empty line after tool executions
    - Thinking blocks now render inline in proper order with text