Change compaction behavior so the compaction entry has no parent (empty
ParentID), creating a new root for post-compaction history. This ensures
old compacted messages are not traversed when building LLM context.
- Modify AppendCompaction to create entries with empty ParentID
- Update BuildContext to collect kept messages via FirstKeptEntryID
- Update GetContextEntryIDs with same logic
- Add comprehensive tests for compaction behavior
- Add web viewer support for displaying compaction entries
Remove dead code now that pkg/kit transparently handles <thinking> and
tags at the agent layer. The ACP server no longer needs to:
- Track inThinkingTag state across chunks
- Parse and split reasoning/text from MessageUpdateEvent chunks
- Maintain tag format constants
MessageUpdateEvent now contains clean text, and ReasoningDeltaEvent
contains structured reasoning - no duplicate filtering needed.
- Add width parameter to UserBlock and apply lipgloss.Wrap() before
passing content to herald Tip alert
- Subtract 4 from width to account for alert bar prefix and margin
- Pass renderer width from RenderUserMessage to UserBlock
- Mirrors the assistant message wrapping added in e33564c
- Replace single Ctrl+S with Ctrl+X leader prefix followed by "s"
- Add leaderKeyActive flag to AppModel for two-key chord state
- Ctrl+X sets the leader flag; next keypress completes or cancels chord
- Update hint text in input component (adjust width thresholds)
- Update /help command output to reflect new keybind
- Add StepMessagesHandler callback to agent's GenerateWithLoopAndStreaming
so callers can persist messages as each step completes
- Wire onStepMessages in Kit.generate() to call session.AppendMessage
for each step's messages immediately on completion
- Track PersistedMessageCount on GenerateWithLoopResult so runTurn
skips already-persisted messages in post-generation cleanup
- Tool calls are always persisted as assistant+tool pairs (never orphaned)
- Document concurrency and incremental persistence requirements on
the SessionManager interface for custom implementations
- ToMarkdown() received a width param but never used it
- Apply lipgloss.Wrap() after herald-md render to break long lines
- Preserves ANSI styles/colors through the wrapping pass
- Fixes overflow for all markdown paths: assistant messages, tool
bodies, and overlay text
- Limit each queued/steering block to 3 visible content lines with ellipsis
- Account for soft-wrapping when counting visual lines
- Truncation is visual only; full text is preserved for scrollback
- Add truncateMessageForBlock helper with wrap-aware line counting
- Add 7 unit tests covering short, exact, overflow, wrapping, and mixed cases
Switch from standard log.Printf to charmbracelet/log for extension loading
messages. This ensures DEBUG output only appears when explicitly enabled.
- Remove unconditional WARN log for failed extension loads
- Convert DEBUG loaded extension message to structured log.Debug call
- Add ToolOutput struct, TextResult/ErrorResult helpers, and
ToolCallIDFromContext so SDK consumers can create custom tools
without importing charm.land/fantasy
- Add NewTool (sequential) and NewParallelTool (concurrent) generic
constructors with automatic JSON schema generation from struct tags
- Remove dead UpdateUsageFromResponse method and fantasy import from
internal/ui/cli.go
- Update SDK skill, README, and www/ docs with custom tool examples
and corrected hook signatures
- Add tea.LogToFile in runInteractiveModeBubbleTea to send stdlib log
output to /tmp/kit/kit.log instead of stderr
- Replace charmbracelet/log with stdlib log in extensions loader,
runner, watcher, prompts loader, and pkg/kit so all log calls go
through the redirected stdlib logger
- Leave charmbracelet/log in CLI-only commands (install, acp) and
acpserver where stderr logging is correct
Add SessionManager interface to allow pluggable session storage backends.
This enables users to implement custom session managers for databases,
cloud storage, or other persistence mechanisms instead of the default
JSONL file-based TreeManager.
Changes:
- Add SessionManager interface with methods for message storage,
tree navigation, compaction, and extension data
- Add treeManagerAdapter to wrap existing TreeManager for backward compatibility
- Update Kit struct to use SessionManager interface instead of concrete type
- Add SessionManager option to Options struct
- Update all session-related methods to use interface
- Add documentation for custom SessionManager usage
The default behavior is preserved - when no SessionManager is provided,
Kit automatically uses the TreeManager via the adapter.
Add two new Options fields for programmatic SDK usage:
- SkipConfig: Skip .kit.yml file loading while still using viper defaults
and environment variables. Useful for fully programmatic configuration.
- DisableCoreTools: Allow creating agents with 0 tools (chat-only mode) or
with only custom tools. When true and Tools is empty, no tools are loaded.
When combined with custom Tools, only those tools are loaded.
Updates documentation in README, pkg/kit/README, skills/kit-sdk/SKILL,
and www/pages/sdk/options.
Remove charmbracelet/log debug statements from the file watcher that
were writing directly to stderr, corrupting the Bubble Tea terminal UI.
- Remove log.Debug calls for directory operations and file changes
- Remove log.Warn for watcher errors (silently ignore instead)
- Remove the charmbracelet/log import entirely
Background MCP tool loading (added in 7e54710) caused tools to not appear
in the UI because tool names and counts were captured at startup before
loading completed. This adds:
- MCPToolsReadyEvent and MCPServerLoadedEvent for progress notifications
- Dynamic GetToolNames/GetMCPToolCount callbacks for live updates
- Per-server status messages as each MCP server finishes loading
- Refresh handlers to update /tools output and status bar when ready
- Detect new subdirectory creation in the fsnotify event loop and add
it to the watcher so files created inside trigger reload events
- Handle cp -r case by checking if new directories already contain
matching files and scheduling an immediate debounced reload
- Add dirContainsMatchingFiles helper method
- Add tests for both new-subdirectory and copy-with-existing-files cases
- Add internal/watcher package with general-purpose ContentWatcher
using fsnotify, configurable file extensions, and debouncing
- Add ContentReloadEvent and App.NotifyContentReload() for TUI signaling
- Add GetPromptTemplates/GetSkillItems callback fields on AppModelOptions
following the existing GetExtensionCommands lazy-provider pattern
- Add Kit.ReloadSkills() to re-discover skills from disk
- Wire fsnotify watcher for .kit/prompts/, .kit/skills/, .agents/skills/,
and global config directories, triggering on .md/.txt changes
- TUI refreshes autocomplete entries and skill list on reload
- Always fire onResponse callback even when response text is empty so
ResponseCompleteEvent reaches the TUI and resets the StreamComponent
- Check for existing StreamingMessageItem in flushStreamAndPendingUserMessages
before creating a new StyledMessageItem to avoid duplicate content
- Mark trailing StreamingMessageItem complete on StepComplete, StepCancelled,
and StepError to freeze live timers and prevent dangling streaming state
Load MCP server tools in the background so the UI appears immediately
instead of blocking until all servers connect. The first LLM call
automatically waits for tools to be ready before proceeding.
Key changes:
- NewAgent() starts MCP loading in a background goroutine and returns
immediately with core/extension tools only
- GenerateWithLoop() calls ensureMCPTools() to lazily wait and rebuild
the fantasy agent with full tool set before first LLM call
- Parallelize LoadTools() across all configured MCP servers
- Add WaitForMCPTools() and MCPToolsReady() for status checking
- Refactor SetModel/SetExtraTools to use shared rebuildFantasyAgent()
- Expose async MCP status methods in public SDK
- Add --frequency-penalty and --presence-penalty CLI flags (0.0-2.0)
- Wire through config, viper, ProviderConfig, and fantasy agent options
- Support in config file, env vars (KIT_FREQUENCY_PENALTY), and SDK
- Pass to Ollama via options map (frequency_penalty, presence_penalty)
- Apply on both initial agent creation and runtime model swap
Add SystemPromptEntry type to capture system prompt, model, and provider
when sharing sessions via /share command. The entry is inserted into the
JSONL after the header and displayed in the web viewer as a collapsible
section with a model badge.
- Add SystemPromptEntry with Content, Model, and Provider fields
- Capture current system prompt and model at share time
- Display in web viewer with collapsible UI and model badge
- Update documentation for /share command
- Add MCPAuthHandler interface at SDK level (pkg/kit/) so all consumers
(CLI, TUI, SDK embedders) control the OAuth UX through one interface
- Default handler opens system browser + local callback server with PKCE
- CLIMCPAuthHandler wraps default with status messages (stderr pre-TUI,
system messages via TUI event system once running)
- Always enable OAuth on remote transports (streamable HTTP, SSE) when
handler is configured; harmless for servers that don't need it
- Dynamic client registration when no client ID is pre-configured
- File-based TokenStore persists tokens to ~/.config/.kit/mcp_tokens.json
keyed by server URL so users don't re-auth on restart
- Catch OAuthAuthorizationRequiredError at connection init (startup) and
tool execution (mid-session token expiry), run auth flow, retry once
- Fix error wrapping (%v -> %w) in connection pool so errors.As can
unwrap through the chain to find OAuth errors
- Thread AuthHandler through MCPToolManager -> AgentConfig ->
AgentCreationOptions -> AgentSetupOptions -> kit.Options
- InstallWithInclude wrote manifest twice via two different code paths,
with the first write missing Include; unify into shared install() method
that writes the manifest once with all fields including Include
- Update() now reads the existing manifest entry to preserve Include and
Installed timestamp instead of constructing a fresh entry from scratch
- Filter out _test.go files in findExtensionsInDir, findExtensionsInRepo,
and ScanForExtensions to prevent Yaegi from loading test files
- Narrow examples/ traversal so only recognized extension directories
(extensions/, ext/, *-ext/, *-extensions/) are scanned, not arbitrary
subdirs like examples/sdk/ that import pkg/kit
- Send sendChatAction("typing") every 4s while agent is processing,
started on AgentStart and stopped on AgentEnd/SessionShutdown
- configPath() now checks project-local .kit/ first, then falls back
to ~/.config/kit/kit-telegram.json for cross-project portability
- ctx.Abort(): cancel current agent turn and clear queue without
injecting a new message (App.Abort + App.IsBusy methods)
- ctx.IsIdle(): check whether the agent is currently processing
- ctx.Compact(CompactConfig): trigger async context compaction with
OnComplete/OnError callbacks (App.CompactAsync method)
- ctx.SendMultimodalMessage(text, []FilePart): send text+image messages
to the agent, bridging ext.FilePart to fantasy.FilePart via RunWithFiles
- ctx.GetSessionUsage() SessionUsage: expose aggregated session token
usage and cost from the UsageTracker
New types: CompactConfig, FilePart, SessionUsage
Wired in both context setups in cmd/root.go with nil-guard defaults
in runner.go and Yaegi symbol exports in symbols.go
Move reasoning tag detection from the provider and UI layers into the agent layer. This prevents raw XML tags from leaking into text streams while ensuring structured reasoning events are emitted correctly for all callers.
- Add `baseUrl` and `apiKey` fields to CustomModelConfig (config and models packages)
- Store them on ModelInfo so they travel through the registry
- createCustomProvider resolves URL/key from model definition first,
falling back to global --provider-url / --provider-api-key
- Fix registry initialisation: call ReloadGlobalRegistry() in InitConfig()
so customModels from config are visible on startup (not just at init time)
- Include custom provider in GetLLMProviders() so custom models appear
in the /model selector
- Hide the built-in custom/custom stub from the selector when user-defined
custom models are present
- Add full-width bordered container with rounded border and primary color
- Add max height constraint to prevent terminal overflow
- Improve selection highlighting with inverted colors matching PopupList style
- Change cursor indicator from › to > for consistency
- Add separator lines between header, content, and footer
- Add footer showing current filter mode
Steering messages (Ctrl+S during agent work) now carry file attachments
just like queued messages do. Previously, pasted images were silently
dropped when steering.
Changes:
- Add SteerMessage struct with Text and Files fields
- Update steer channel from chan string to chan SteerMessage
- Add SteerWithFiles methods through the stack (UI, app, SDK)
- Update PrepareStep to include files in injected user messages
- Add full-width bordered container with rounded border and primary color
- Add max height constraint to prevent terminal overflow
- Improve selection highlighting with inverted colors matching PopupList style
- Change cursor indicator from › to > for consistency
- Use MutedBorder for tree lines and Success color for active marker
- Update search display format to match PopupList (
Herald's codeBlockWithLineNumbers() hardcodes PaddingTop(1) and
PaddingBottom(1), adding invisible blank lines with background color
above and below the code content. These padding lines occupy line
indices in the rendered item but are visually indistinguishable from
empty space, causing mouse click coordinates to map to the wrong
content line (consistently 1 row off in tool output blocks).
Strip the padding lines after CodeBlock rendering since the Compose
separator above and Figure caption below already provide adequate
visual spacing.
Add dedicated renderGrepBody function for the grep tool, replacing the
previous behavior of routing it through renderBashBody. The grep tool now:
- Shows a caption with total match count (e.g., '8 matches' or '1 match')
- Displays truncation info when matches exceed maxLsLines
- Uses consistent Figure component styling with ls, read, find, and bash tools
- Uses 'match/matches' terminology appropriate for grep results
Add dedicated renderFindBody function for the find tool, replacing the
previous behavior of routing it through renderBashBody. The find tool now:
- Shows a caption with total result count (e.g., '12 results')
- Displays truncation info when results exceed maxLsLines
- Uses consistent Figure component styling with ls, read, and bash tools
Apply the same Figure component pattern to the ls tool for consistency
with read and bash tools. The caption now appears below the directory
listing and shows the count of hidden entries when truncated.
Replace inline truncation hints and exit code labels with herald's
Figure component. Captions now appear below content and show:
- read: filename • lines X-Y of Z • offset=N to continue
- bash: N more lines • exit code N
This provides consistent visual grouping and cleaner metadata
display for tool output blocks.
- Replace detachedWithCancel (goroutine-based) with context.WithoutCancel
+ valuesContext; the old goroutine would fire immediately if the parent
was already cancelled/deadline-exceeded, causing 'failed after 0s'
- Kit.Subagent() pre-flight: if the incoming ctx is already done, reset
to context.Background() before applying the subagent timeout
- Both Subagent() error paths now return a non-nil *SubagentResult with
Elapsed set, so the tool response always shows accurate timing
- Narrow viperInitMu scope in Kit.New(): snapshot viper state + call
BuildProviderConfig under the lock, then release before SetupAgent /
MCP loading; parallel subagent spawns no longer serialise on viper I/O
- AgentSetupOptions gains ProviderConfig + scalar fields so SetupAgent
can skip viper reads when a pre-built config is supplied
- Add subagent_test.go covering the fixed context detachment behaviour
- Add fsnotify-based file watcher that auto-reloads extensions on .go
file changes in autoloaded dirs with 300ms debounce
- Add /reload-ext built-in command (alias /re) for manual reload
- Add Agent.SetExtraTools() so extension tools update on reload
instead of being baked in at agent creation time
- Run reload async via tea.Cmd to avoid prog.Send() deadlock when
extension handlers call ctx.Print() during SessionStart/Shutdown
- Wire watcher lifecycle into cmd/root.go with graceful shutdown
- Break scanner bar into individual LED segments with single-space gaps
- Center KIT text over the scanner bar (13-space indent for all lines)
- Maintain original 46-char total width for the scanner bar
- Add ValidateModelString() to ModelsRegistry for format, provider,
and model name validation with typo suggestions
- Validate model in Kit.Subagent() before expensive Kit.New() setup
- Remove silent fallback to parent model on creation failure
- Error propagates as tool result so calling agent can self-correct
- Add registry_test.go covering format, provider, and suggestion cases
- Add PopupList: generic themed popup with fuzzy search, scrolling,
keyboard navigation, and centered overlay rendering
- Refactor ModelSelectorComponent to delegate to PopupList instead
of implementing its own full-screen rendering and input handling
- Render /model selector as a centered overlay on top of the chat
view instead of replacing the entire screen
- PopupList accepts a pluggable FilterFunc for domain-specific
fuzzy matching (model selector wires its own scoring)
- Add 11 tests for PopupList covering navigation, search, selection,
cancellation, filtering, rendering, and edge cases
- Wire fantasy's OnReasoningEnd callback through the full event chain:
agent → SDK (ReasoningCompleteEvent) → app → TUI
- Freeze reasoning duration in both StreamComponent and
StreamingMessageItem as soon as reasoning ends, not when the
next assistant text chunk arrives
- Fix accent color on duration label in render.ReasoningBlock to
match the live streaming style (VeryMuted prefix + Accent duration)
- Refactor go-edit-lint to collect edited .go files during the agent
turn via OnToolResult, then run gopls + golangci-lint once in
OnAgentEnd instead of after every individual edit/write call
- Use ctx.SendMessage() to inject diagnostics as a follow-up prompt
when issues are found, replacing the old tool-result rewriting
- Show a green 'all clean' block when no issues are detected
- Fix StopReason docs in skills/kit-extensions/SKILL.md: the value is
'error' on failure, 'completed' when the LLM returns empty, or the
raw provider value (e.g. 'stop', 'end_turn') passed through — not
the previously documented 'completed'/'cancelled'/'error' enum
Remove 8 unused exports from clipboard package:
- CopyToClipboardWithMessage, IsClipboardSupported
- ToastMsg, ToastType, ToastInfo, ToastSuccess, ToastWarning, ToastError
These were remnants of a toast notification feature that was never
wired up. No callers exist anywhere in the codebase.
Ed Zynda