- Use max() instead of if statement for min value
- Use strings.SplitSeq for efficient iteration
- Use range over int instead of explicit loop counter
- Remove unused functions:
- InputComponent.renderPopup()
- AppModel.renderStream()
- AppModel.renderStreamingBashOutput()
- AppModel.printCompactResult()
flushStreamContent() was creating a new StyledMessageItem when tool calls
started, but we already had a StreamingMessageItem with the same content.
Now we:
- Mark the existing StreamingMessageItem as complete
- Only create a new message as fallback if no streaming item exists
This fixes text duplication when assistant messages precede tool calls.
- Limit streaming bash output to 20 lines max during live display
- Remove streaming bash item when tool completes
- Replace with truncated tool result block
- Expand background color to full terminal width with proper indentation
- Matches renderBashBody styling (lineIndent + width)
This prevents long-running commands from growing the UI forever while
still showing live output up to a reasonable height.
- Add StreamingBashOutputItem to message_items.go
- Update ToolOutputEvent handler to append chunks to bash item in ScrollList
- Remove old renderStreamingBashOutput() that broke layout
- Bash output now streams inline with messages instead of separate section
- Auto-scrolls to bottom during streaming
- Marks bash item complete on ToolResultEvent
Fixes layout breaking when bash commands produce streaming output.
- Add StreamCallback parameter to compaction.Compact() for streaming text deltas
- Update generateSummary() to use fantasy.Agent.Stream() when callback provided
- Fix compactSplitTurn() to stream both history and turn prefix summaries
- Add SDK event subscription in CompactConversation() goroutine
- Update UI to handle streaming compaction like regular assistant messages
- Compaction summaries now stream word-by-word instead of appearing all at once
Fixes issue where compaction would show incomplete context (e.g. only 'nce')
by ensuring both history summary and turn prefix are streamed to the UI.
- Move popup rendering from inline (below input) to centered overlay
- Add RenderPopupCentered() method to InputComponent
- Implement overlayContent() helper for line-by-line merging
- Popup now appears in center of screen above all content
- Prevents overflow issues when typing / at bottom of terminal
- Detect role transition in appendStreamingChunk (reasoning → assistant)
- Mark reasoning StreamingMessageItem as complete when assistant text starts
- Duration counter now freezes immediately when reasoning ends
- Add live duration counter that updates during reasoning streaming
- Store startTime and finalDuration for proper counter behavior
- Create StreamingMessageItem that accumulates chunks and re-renders
- Update StreamChunkEvent/ReasoningChunkEvent to append to StreamingMessageItem
- Enable live streaming display within ScrollList (iteratr-style)
- Mark streaming items as complete on ResponseCompleteEvent
- Reasoning and assistant text now stream in real-time in the viewport
Disable the mouse selection and keyboard copy features while keeping
all the supporting code infrastructure:
- Comment out MouseClickMsg, MouseMotionMsg, MouseReleaseMsg handlers
- Comment out keyboard shortcuts (c/y keys) for copying
- Keep all ScrollList selection tracking code
- Keep clipboard utilities (clipboard.go)
- Keep highlighting functions in scrolllist.go
This allows the features to be easily re-enabled later while keeping
the codebase clean for now.
Implement visual feedback for text selection in the scrollback:
- Add isLineInSelection() to check if a line is within the current selection
- Add applyHighlight() using the theme's Highlight color for selected lines
- Add applyFocusIndicator() using MutedBorder for focused items
- Update View() to apply highlighting during rendering
- Add getItemAndLineAtY() for precise mouse position tracking
- Track both item index and line index within item for selection
Selection highlighting uses the user's selected theme colors for
consistent visual feedback across all themes.
Implement mouse selection and keyboard copy functionality following
crush's patterns:
- Add clipboard.go with dual-write clipboard support (OSC 52 + system)
- Add CopySelection tracking to ScrollList for text selection
- Implement HandleMouseDown/HandleMouseDrag/HandleMouseUp methods
- Add keyboard shortcuts (c/y) for copying messages
- Mouse click+drag to select text, auto-copy on release
- Toast notifications for copy feedback
Note: Full text extraction from selection requires additional work to
properly extract raw text from styled message content.
Add SetSessionModel method to the ACP agent, allowing clients to change
the active LLM model for a session at runtime. The method looks up the
session in the registry and delegates to kit.SetModel().
Verified with smoke test: session/set_model now returns success instead
of 'Method not found' error.
Replace concrete LLMMessage/LLMUsage/LLMResponse/LLMFilePart structs with
type aliases to charm.land/fantasy types, exposing them under clean
LLM-prefixed names. This gives SDK consumers full access to rich message
parts (tool calls, reasoning, tool results) without importing fantasy
directly.
Key changes:
- LLM types are now aliases: LLMMessage=fantasy.Message, etc.
- Added aliases for all part types: LLMTextPart, LLMToolCallPart, etc.
- Re-exported constructors: NewLLMUserMessage, NewLLMSystemMessage
- Removed lossy conversion helpers (llm_convert.go, fantasyMsgsToKit)
- Updated all internal packages to use aliases consistently
- Added ACP smoke test script and prompt template
- Fixed lint issues: unused vars, modernize min() usage
Remove charm.land/fantasy from the public API surface of pkg/kit by
replacing the four type aliases with concrete Kit-owned structs:
- LLMMessage {Role LLMMessageRole, Content string}
- LLMUsage {InputTokens, OutputTokens, TotalTokens, ...}
- LLMResponse {Content, FinishReason, Usage}
- LLMFilePart {Filename, Data []byte, MediaType}
Add LLMMessageRole type with user/assistant/system/tool constants.
Introduce pkg/kit/llm_convert.go as the single boundary layer where
Kit types convert to/from fantasy types internally. All callers in
pkg/kit, pkg/kit/compaction.go, pkg/kit/extensions_bridge.go, and
internal/app/app.go cross through this layer.
ContextPrepareHook.Messages and ContextPrepareResult.Messages change
from []fantasy.Message to []LLMMessage. extensions_bridge.go drops
its fantasy and strings imports entirely.
internal/app/app_test.go switches &fantasy.Usage{} to &kit.LLMUsage{}.
Add seven new tests in types_test.go covering concrete construction,
role constants, JSON snake_case tags, and round-trip conversion.
- Remove always-zero queueLen variable: len() was measured after
clearing the queue, so it was unconditionally 0 and the variable
was dead code
- Emit QueueUpdatedEvent{Length: 0} explicitly to make intent clear
- Also emit QueueUpdatedEvent when a second batch is pulled mid-loop;
previously the queue was silently cleared without notifying the UI,
leaving queuedMessages stuck in the displayed-queued state forever
/new-prompt <description> scaffolds a new .kit/prompts/ template.
Explains the file format, argument substitution syntax, naming
conventions, and writing guidelines.
Provides a /commit-push slash command that reviews git status and diff,
stages all changes, writes a conventional commit message, commits, and
pushes to the current branch.
The compact display mode was purely a UI concern that added complexity
without providing unique value. Anyone wanting compact-style formatting
can implement it as an extension using the Renderer interface.
- Delete internal/ui/compact_renderer.go
- Remove renderToolBodyCompact and all compact tool body renderers from
tool_renderers.go
- Simplify NewCLI(debug bool) — drop compact parameter
- Simplify NewStreamComponent(width, modelName) — drop compactMode parameter
- Remove CompactMode from AppModelOptions, app.Options, CLISetupOptions
- Remove Compact from internal/config/config.go
- Remove --compact flag, var, and viper binding from cmd/root.go
- Update format.go: remove CompactRenderer interface compile-time check
and clean up comments
When using /resume to resume a session, reasoning/thinking content
was not being displayed even though it was saved in the session file.
Changes:
- Add RenderReasoningBlock to Renderer interface
- Implement RenderReasoningBlock for MessageRenderer with muted italic
styling matching live streaming output
- Implement RenderReasoningBlock for CompactRenderer with same styling
- Update renderSessionHistory to render reasoning content before
assistant message text
Fixes: reasoning blocks now populate correctly when resuming sessions
The Qwen model outputs thinking content wrapped in tags
(not <thinking>). Updated parseThinkingTags to detect and handle
both formats:
- <thinking>...</thinking> (long format)
- (short format)
Also removed the hasProperReasoningEvents logic that was preventing
thinking tag parsing from working correctly. Now both ReasoningDeltaEvent
(from models with proper reasoning APIs) and thinking tags in text
(from models like Qwen) are handled together, matching the TUI behavior.
Track whether a model sends proper ReasoningDeltaEvent events. If so,
skip parsing <thinking> tags from text to avoid sending reasoning content
twice (once as proper reasoning, once parsed from text).
Also reset the tracking state at the start of each new prompt turn.
Parse <thinking>...</thinking> tags from models (Qwen, DeepSeek) that
wrap reasoning content in XML-style tags instead of using proper
reasoning events.
When text chunks contain thinking tags:
- Extract content between tags and send as reasoning/thought updates
- Send content outside tags as regular message text
- Track state across chunks to handle streaming properly
This mirrors the TUI's thinking tag parsing behavior.
- Implement proper handling for all ACP content block types:
- ContentBlockText: extracts text content
- ContentBlockImage: decodes base64 to LLMFilePart
- ContentBlockAudio: decodes base64 to LLMFilePart
- ContentBlockResource: handles text and binary embedded resources
- ContentBlockResourceLink: reads files from disk
- Text files are now included inline in the message (not as FilePart)
to avoid OpenAI API errors. Only binary files (images, audio, PDFs)
are sent as FilePart attachments.
- Add fallback MIME types when not provided by client
- Add default prompt text when user attaches files without text
- Add comprehensive debug logging for content extraction
- Enable debug logging in ACP command when --debug flag is used
Previously, when streaming text grew taller than the allocated view
height, the top (older) lines were silently discarded by viewContent().
This meant users could not scroll up to see them.
Now, overflow lines are emitted directly via tea.Println so they land
in the terminal's real scrollback buffer — matching the diagram where
completed text lives in the red scrollback region and the green viewable
area always shows the most recent streaming lines + input/footer.
Key changes:
- StreamComponent: add scrollbackFlushedLines counter and ConsumeOverflow()
method that returns newly overflowed lines and advances the pointer
- StreamComponent.Reset(): zero the counter between steps
- StreamComponent.GetRenderedContent(): skip already-flushed lines so
the end-of-step flush doesn't re-emit content already in scrollback
- AppModel.Update(): call ConsumeOverflow() each cycle and emit overflow
directly via tea.Println (not appendScrollback, to avoid triggering
drainScrollback's auto-flush guard while streaming is active)
- streamComponentIface: add ConsumeOverflow() to interface
- model_test.go: add stub ConsumeOverflow() to test double
- children_test.go: add 7 unit tests covering ConsumeOverflow and the
updated GetRenderedContent skip-flushed-lines behaviour
When users run `/theme <name>`, the alert colors (Tip, Note, Warning, etc.)
now update correctly. Previously, MessageRenderer and StreamComponent cached
herald.Typography instances that weren't refreshed after theme changes.
Changes:
- Added UpdateTheme() method to Renderer interface
- Implemented UpdateTheme() for MessageRenderer to recreate herald typography
- Added no-op UpdateTheme() stub for CompactRenderer (fetches colors fresh)
- Implemented UpdateTheme() for StreamComponent reasoning block renderer
- Modified handleThemeCommand() to notify all renderers of theme changes
This ensures newly rendered messages use the current theme's alert colors.
- Display kitBanner() before PrintStartupInfo() when running Kit normally
- The ASCII art banner with KITT scanner lights now appears at the top
of the screen, before Model, Context, Skills information
- Maintains consistent styling with the existing usage/help screen
Models like Qwen and DeepSeek wrap reasoning content in ... XML-like
tags within the regular content field. This was causing the reasoning
text to appear twice - once as a reasoning block and once as regular text.
Changes:
1. Provider hooks (providers.go):
- Extract reasoning from tags and emit proper reasoning events
- Use openai provider directly with custom ExtraContentFunc and
StreamExtraFunc hooks to parse thinking content
2. Stream filtering (stream.go):
- Filter out all text content between and tags at the
streaming level to prevent duplicate rendering
- Track state with inThinkTag flag across stream chunks
3. Message conversion (content.go):
- Strip any remaining tags from text content when converting
from fantasy messages
The regex patterns use string concatenation to avoid XML tag corruption:
regexp.MustCompile( + + + + + + + )
Fixes duplicate reasoning text when using custom provider with models
that wrap thinking in tags.
Two related fixes for --provider-url handling:
1. Don't restore custom/* models from preferences without --provider-url
- When user runs with --provider-url, model defaults to custom/custom
- If they switch models, custom/custom gets persisted to preferences
- On next run without --provider-url, restoring custom/custom fails
- Now we skip restoring custom/* models when no --provider-url is provided
2. Auto-prefix bare model names with custom/ when --provider-url is set
- Users often provide just the model name (e.g., qwen3.5-35b-a3b)
- This failed with 'invalid model format' error
- Now auto-prefixed with custom/ for OpenAI-compatible endpoints
- Replace glamour-based markdown rendering with herald/herald-md
- Update go.mod and go.sum with new dependencies
- Refactor styles.go to use Typography cache instead of TermRenderer
- Update enhanced_styles.go for compatibility
- Update btca.config.jsonc configuration
- Update github.com/indaco/herald v0.9.0 -> v0.10.0
- Update charm.land/bubbles/v2 v2.0.0 -> v2.1.0
- Update AWS SDK v2 packages
- Update google.golang.org/genai v1.51.0 -> v1.52.0
- Update various other dependencies
refactor(ui): use herald.CodeBlock for Read tool output
- Replace manual renderCodeBlock() with herald.CodeBlock()
- Add WithCodeLineNumberOffset() support for correct line numbers
- Extract language hint from file extension for syntax highlighting
- Preserve existing syntax highlighting via WithCodeFormatter()
- Remove unused codeLine struct and renderCodeBlock function
Add two test files that auto-discover and validate every single-file
extension in examples/extensions/:
- all_extensions_load_test.go: Verifies all 32 extensions load into the
Yaegi interpreter without errors (syntax, imports, Init signature).
- all_extensions_sanity_test.go: Six generalized sanity checks:
- Lifecycle: SessionStart → SessionShutdown round-trip
- CommandSanity: non-empty names/descriptions, no spaces/leading slash,
non-nil Execute, no duplicates
- ToolSanity: non-empty names/descriptions, at least one executor,
valid JSON parameters, no duplicates
- ZeroValueEvents: all 22 event types fired as zero-value structs
- WidgetSanity: non-empty IDs, consistent keys, valid placements
- IdempotentLifecycle: repeated SessionStart/SessionShutdown
Shared extensionFiles() helper auto-discovers extensions so new files
are automatically covered.
performFork() called ClearMessages() after Branch(targetID), but
ClearMessages() calls TreeSession.ResetLeaf() which sets leafID back
to empty — immediately undoing the branch. The in-memory message store
was also never reloaded from the tree session after branching, so the
LLM had zero context.
Add ReloadMessagesFromTree() which clears the store and reloads it
from the tree session's current branch without resetting the leaf
pointer. Use it in performFork() instead of ClearMessages().
The issue was that cache control persisted across turns in conversation
history, causing accumulation beyond Anthropic's 4-block limit.
Changes:
- Count existing cache blocks in message history before adding new ones
- Only add new cache blocks up to the 4-block limit
- Remove tool caching (was adding 1 block per turn)
- Skip messages that already have cache control set
Tested with 5 sequential messages - no errors, proper cache metrics.
Crush's proven 4-block strategy:
1. Last system message (if present)
2. Last 2 conversation messages
3. Last tool definition
This stays exactly at Anthropic's 4-block limit without exceeding it.
Previous implementation could exceed the limit in certain edge cases.
Now matches Crush's battle-tested approach.
Anthropic API enforces a maximum of 4 blocks with cache_control per request.
The previous implementation could exceed this limit when combining:
- System message caching
- Recent message caching
- Tool definition caching
Changes:
- Add explicit cache block counting (max 4)
- Remove tool cache control to stay under limit
- Prioritize: system message first, then recent messages
- Work backwards from end to cache most recent context first
Fixes: bad request error 'A maximum of 4 blocks with cache_control may be provided'
- Fix gofmt formatting issues in 7 files
- Replace atomic.AddUint64 with atomic.Uint64 type (modernize)
- Replace for i := 0; i < count; i++ with for i := range count (modernize)
- Replace strings.Split with strings.SplitSeq (modernize)
- Replace deprecated GetFantasyProviders with GetLLMProviders
- Replace deprecated GetFantasyMessages with GetLLMMessages
- Replace deprecated ConvertFromFantasyMessage with ConvertFromLLMMessage
- Replace deprecated FromFantasyMessage with FromLLMMessage
- Replace deprecated ToFantasyMessages with ToLLMMessages
- Remove 2 unused formatToolArgs functions
- Upgrade golangci-lint to v2.11.4
- Fix errcheck warnings for os.Setenv/os.Unsetenv in tests
- Use maps.Copy instead of manual loop (modernize lint)
- Add maps import for maps.Copy
Remove internal monologue comments that don't add value for readers:
- Remove lengthy explanations of type conflicts that are now resolved
- Remove 'NOTE:' and 'TODO:' comments documenting implementation history
- Remove obvious test comments that just restate what the code does
- Keep only meaningful comments that explain design intent
The code is now cleaner and easier to read without the self-referential
commentary that was useful during development but not for maintenance.
Implements automatic prompt caching to reduce API costs by 60-90% for
repeated prompts with the same context.
Architecture:
- Provider-level caching for OpenAI (PromptCacheKey)
- Message-level caching for Anthropic (avoids type conflicts)
- Model family detection enables caching regardless of provider
Key Changes:
- Add ModelInfo.Family with SupportsCaching() and CacheType() methods
- Add ProviderConfig.DisableCaching for opt-out
- Implement message-level cache control in agent (like Crush)
- Last system message gets cache control
- Last 2 messages get cache control
- Last tool gets cache control
- Auto-disable caching when thinking is enabled (type conflict avoidance)
- Add KIT_DISABLE_CACHE environment variable for global opt-out
Tested with opencode/claude-sonnet-4-6 showing cacheRead/cacheWrite
tokens in debug output, confirming 60-90% cost savings.
Closes cost optimization for multi-turn conversations.
Rename public SDK symbols to use generic LLM terminology instead of
exposing the internal dependency name (charm.land/fantasy):
Public API renames (with deprecated wrappers for backward compat):
- ConvertToFantasyMessages() → ConvertToLLMMessages()
- ConvertFromFantasyMessage() → ConvertFromLLMMessage()
- GetFantasyProviders() → GetLLMProviders()
New type alias:
- LLMFilePart = fantasy.FilePart (eliminates need for direct fantasy import)
- PromptResultWithFiles() signature now uses LLMFilePart
Internal renames (with deprecated wrappers):
- ModelsRegistry.GetFantasyProviders() → GetLLMProviders()
- TreeManager.GetFantasyMessages() → GetLLMMessages()
- TreeManager.AppendFantasyMessage() → AppendLLMMessage()
- TreeManager.AddFantasyMessages() → AddLLMMessages()
- Message.ToFantasyMessages() → ToLLMMessages()
- FromFantasyMessage() → FromLLMMessage()
- npmToFantasyProvider → npmToLLMProvider
- isProviderFantasySupported() → isProviderLLMSupported()
All internal callers migrated to new names. ~30 comments updated
to remove Fantasy references across pkg/kit/, internal/agent/,
internal/models/, internal/message/, internal/session/.
Documentation updates:
- AGENTS.md: added Public SDK rules section (no dependency leakage,
naming conventions, deprecation pattern)
- README.md: removed Fantasy references
- pkg/kit/README.md: full rewrite with current API surface
- skills/kit-sdk/SKILL.md: updated examples and type references
- www/pages/providers.md, www/pages/cli/commands.md: updated
Ed Zynda