Breaking changes:
- Agent class moved to @mariozechner/pi-agent-core
- Transport abstraction removed (ProviderTransport, AppTransport)
- AppMessage renamed to AgentMessage
- UserMessageWithAttachments now has role: 'user-with-attachments'
- CustomMessages interface replaced with CustomAgentMessages
- agent.appendMessage() removed (use queueMessage())
- New agent event types from pi-agent-core
Added:
- defaultConvertToLlm, convertAttachments utilities
- isUserMessageWithAttachments, isArtifactMessage type guards
- createStreamFn for CORS proxy support
- Default streamFn and getApiKey on AgentInterface
- Proxy utilities exported
README completely rewritten to reflect new architecture.
- Export wrapTextWithAnsi from tui package
- Add 'Handling Input' section to README with key detection example
- Document wrapTextWithAnsi in utilities section
- Add selectedBg theme color for active line highlight
- Fix filter modes:
- no-tools: default minus tool results (still hides label/custom)
- user-only: just user messages
- labeled-only: just labeled entries
- all: everything
- Update theme.md with new color tokens (50 total)
- Add GoogleThinkingLevel type mirroring Google's ThinkingLevel enum
- Update GoogleGeminiCliOptions and GoogleOptions to use our type
- Cast to any when assigning to Google SDK's ThinkingConfig
- New utils.ts with shared functions:
- FileOperations type and createFileOps()
- extractFileOpsFromMessage()
- computeFileLists()
- formatFileOperations()
- serializeConversation()
- SUMMARIZATION_SYSTEM_PROMPT
- branch-summarization.ts now uses:
- Serialization approach (conversation as text, not LLM messages)
- completeSimple with system prompt
- Shared utility functions
- serializeConversation now takes Message[] (after convertToLlm)
- Handles all custom message types via convertToLlm
- Includes thinking blocks as [Assistant thinking]
- Removes truncation of tool args and results (already token-budgeted)
Instead of passing conversation as LLM messages (which makes the model
try to continue it), serialize to text wrapped in <conversation> tags.
- serializeConversation() formats messages as [User]/[Assistant]/[Tool result]
- Tool calls shown as function(args) format
- Tool results truncated to prevent bloat
- Conversation wrapped in <conversation> tags in the prompt
- SUMMARIZATION_SYSTEM_PROMPT explains the task clearly
- Tells model to NOT continue conversation, ONLY output summary
- Updated prompts to reference 'messages above' for clarity
- Pass systemPrompt to completeSimple
- CompactionDetails type with readFiles/modifiedFiles
- extractFileOperations collects from tool calls and previous compaction details
- UPDATE_SUMMARIZATION_PROMPT for merging with previous summary
- generateSummary now accepts previousSummary for iterative updates
- compact() extracts files, passes previousSummary, returns details
- Only merges from !fromHook compaction entries (backward compatible)
Both auto_compaction_end and executeCompaction were manually adding
CompactionSummaryMessageComponent instead of using addMessageToChat,
which caused missing spacers. Now both use addMessageToChat for
consistent spacing and expansion state handling.
Prepends: 'The user explored a different conversation branch before
returning here. Summary of that exploration:'
This helps the LLM understand the summary is background context from
a different path, not the current thread being continued.
The old messagesToText only extracted text content, losing all tool calls.
Now uses convertToLlm like compaction does, preserving the full conversation
including tool calls, which is essential for generating useful summaries.
getHookPaths(), getCustomToolPaths(), and getSkillsSettings() now return
copies of arrays instead of references to internal state. This prevents
callers from accidentally mutating settings without going through setters.
Fixes#361
Both now use consistent sections:
- Goal
- Constraints & Preferences
- Progress (Done/In Progress/Blocked)
- Key Decisions
- Next Steps
- Critical Context (compaction only)
Prompts instruct LLM to use EXACT format and preserve file paths/function names.
- BranchPreparation now uses AgentMessage[] instead of custom type
- Reuse getMessageFromEntry pattern from compaction.ts
- Add BranchSummarySettings with reserveFraction to settings.json
- Add getBranchSummarySettings() to SettingsManager
- Use settings for reserveFraction instead of hardcoded value
- Add ReadonlySessionManager interface to session-manager.ts
- Re-export from hooks/index.ts
- Add collectEntriesForBranchSummary() to extract entries for summarization
- Don't stop at compaction boundaries (include their summaries as context)
- Add token budget support to prepareBranchEntries()
- Walk entries newest-to-oldest to prioritize recent context
- Use options object for generateBranchSummary()
- Handle compaction entries as context summaries
- Export new types: CollectEntriesResult, GenerateBranchSummaryOptions
- Move compaction.ts to src/core/compaction/compaction.ts
- Extract branch summarization to src/core/compaction/branch-summarization.ts
- Add index.ts to re-export all compaction utilities
- Update all imports across the codebase
- Add test/utilities.ts with shared helpers (API_KEY, userMsg, assistantMsg, createTestSession)
- Add agent-session-tree-navigation.test.ts with e2e tests for tree navigation
- Add getChildren() method to SessionManager
- Add summaryEntry to navigateTree return type
- Update existing tests to use shared utilities
- Check response.stopReason instead of catching errors for abort detection
- Return result object from _generateBranchSummary instead of throwing
- Fix summary attachment: attach to navigation target position, not old branch
- Support root-level summaries (parentId=null) in branchWithSummary
- Remove setTimeout hack for re-showing tree selector on abort
- Show ├─ for non-last siblings, └─ for last sibling
- Show │ gutter for descendants of non-last siblings
- Properly handle multiple roots display shift
- At indent 0: stay flat unless parent branches
- At indent 1: children go to indent 2 (visual grouping)
- At indent 2+: stay flat for single-child chains
- Remove gutter/connector complexity for now
- Only indent when parent has siblings (branch point)
- gutterLevels array tracks which levels need │ vs spaces
- connector: none/branch/last for ├─/└─ display
- Tool results show formatted call info like [read: path:1-10], [bash: cmd]
- Skip assistant messages with stopReason error/aborted
- Skip assistant messages with only tool calls (no text)
- Format built-in tools like tool-execution.ts (read, write, edit, bash, grep, find, ls)
- Custom tools show truncated JSON args
- Label editing with 'l' key
- Ctrl+O cycles through filters: default -> user-only -> labeled-only -> all
- Preserve cursor position when toggling filters
- Labels shown at front of node
- Normalize tabs/newlines in content
- extractContent limited to 200 chars for performance
- truncateToWidth on all rendered lines
- Iterative flattenTree to avoid stack overflow
- Linear chain optimization (no indent for single-child chains)
- Add TreeSelectorComponent with ASCII tree visualization
- Add AgentSession.navigateTree() for switching branches
- Add session_before_tree/session_tree hook events
- Add SessionManager.resetLeaf() for navigating to root
- Change leafId from string to string|null for consistency with parentId
- Support optional branch summarization when switching
- Update buildSessionContext() to handle null leafId
- Add /tree to slash commands in interactive mode
Major changes:
- Replace monolithic SessionEvent with reason discriminator with individual
event types: session_start, session_before_switch, session_switch,
session_before_new, session_new, session_before_branch, session_branch,
session_before_compact, session_compact, session_shutdown
- Each event has dedicated result type (SessionBeforeSwitchResult, etc.)
- HookHandler type now allows bare return statements (void in return type)
- HookAPI.on() has proper overloads for each event with correct typing
Additional fixes:
- AgentSession now always subscribes to agent in constructor (was only
subscribing when external subscribe() called, breaking internal handlers)
- Standardize on undefined over null throughout codebase
- HookUIContext methods return undefined instead of null
- SessionManager methods return undefined instead of null
- Simplify hook exports to 'export type * from types.js'
- Add detailed JSDoc for skipConversationRestore vs cancel
- Fix createBranchedSession to rebuild index in persist mode
- newSession() now returns the session file path
Updated all example hooks, tests, and emission sites to use new event types.
When custom renderer returns a component with its own background,
don't wrap it in another background box. The nested bg resets
were causing padding to lose its background color.
- agentLoop now accepts AgentMessage[] instead of single message
- agent.prompt() accepts AgentMessage | AgentMessage[]
- Emits message_start/end for each message in the array
- AgentSession.prompt() builds array with hook message + user message
- TUI now receives events for before_agent_start injected messages
Hook commands now always execute immediately, even during streaming.
If a command needs to interact with the LLM, it uses pi.sendMessage()
which handles queueing automatically.
This simplifies the API and eliminates the issue of queued slash
commands being sent to the LLM instead of executing.
- Add BeforeAgentStartEvent and BeforeAgentStartEventResult types
- Add emitBeforeAgentStart to HookRunner
- Call in AgentSession.prompt() before agent.prompt()
- Hook can return a message to inject into context (persisted + visible)
- Add test hook demonstrating custom message rendering and before_agent_start
- CompactionSummaryMessageComponent now extends Box, uses customMessageBg
- New BranchSummaryMessageComponent for branch summaries
- Both use same background color as HookMessageComponent for consistency
- Added Spacer before compaction/branch components in chat
- addMessageToChat: exhaustive switch for all AgentMessage roles
- renderSessionContext: delegates to addMessageToChat, special handling for assistant tool calls and tool results
- export-html formatMessage: exhaustive switch for all AgentMessage roles
- Removed isHookMessage, isBashExecutionMessage type guards in favor of role checks
- Fixed imports and removed unused getLatestCompactionEntry
Two bugs:
1. createCustomMessage was returning role: 'user' instead of preserving
the hook message structure (role: 'hookMessage', customType, etc.)
2. rebuildChatFromMessages wasn't clearing the container before rebuilding
Mario Zechner