- read: actionable notices with offset for continuation
- First line > 30KB: return empty + bash command suggestion
- Hit limit: '[Showing lines X-Y of Z. Use offset=N to continue]'
- bash: tail truncation with temp file
- Notice includes line range + temp file path
- Edge case: last line > 30KB shows partial
- grep: pre-truncate match lines to 500 chars
- '[... truncated]' suffix on long lines
- Notice for match limit and line truncation
- find/ls: result/entry limit notices
- '[N results limit reached. Use limit=M for more]'
- All notices now in text content (LLM sees them)
- TUI simplified (notices render as part of output)
- Never return partial lines (except bash edge case)
- Add truncate.ts utility with truncateHead/truncateTail functions
- Both respect 2000 line and 30KB limits (whichever hits first)
- read: head truncation, returns truncation info in details
- bash: tail truncation, writes full output to temp file if large
- grep: head truncation + 100 match limit
- find: head truncation + 1000 result limit
- ls: head truncation + 500 entry limit
- tool-execution.ts displays truncation notices in warning color
- All tools return clean output + structured truncation in details
- Added totalTokens field to Usage interface in pi-ai
- Anthropic: computed as input + output + cacheRead + cacheWrite
- OpenAI/Google: uses native total_tokens/totalTokenCount
- Fixed openai-completions to compute totalTokens when reasoning tokens present
- Updated calculateContextTokens() to use totalTokens field
- Added comprehensive test covering 13 providers
fixes#130
- Updated isBunBinary detection to check for %7EBUN (URL-encoded ~BUN)
- Simplified build:binary script to remove Unix-specific shell syntax
- Binary now correctly locates supporting files next to executable on Windows
Instead of blocking startup for up to 1 second waiting for the version check,
run it in the background and insert the notification into chat when it completes.
- Browse previously submitted prompts using Up/Down arrow keys
- History is session-scoped and stores up to 100 entries
- Load history from session on continue/resume
- Includes 15 tests for history navigation
fixes#121
Browse previously submitted prompts using Up/Down arrow keys, similar to
shell history and Claude Code's prompt history feature.
- Up arrow when editor is empty: browse to older prompts
- Down arrow when browsing: return to newer prompts or clear editor
- Cursor movement within multi-line history entries supported
- History is session-scoped, stores up to 100 entries
- Consecutive duplicates are not added to history
Includes 15 new tests for history navigation behavior.
stripAnsi misses some escape sequences like standalone ESC \ (String
Terminator) which caused rendering issues when displaying captured TUI
output. Now also removes any remaining ESC+char sequences and control
characters after stripAnsi processing.
- Use truncateToWidth instead of substring in user-message-selector.ts
- Fix truncateToWidth to use Intl.Segmenter for proper grapheme handling
- Add tests for Unicode truncation behavior
- Add 'authHeader' boolean option to models.json provider config
When true, adds 'Authorization: Bearer <apiKey>' to model headers
Useful for providers requiring explicit auth headers (fixes#81)
- Fix print mode (-p) silently failing on errors
Now outputs error message to stderr and exits with code 1
when assistant message has stopReason of error/aborted
- RPC mode now auto-compacts when context exceeds threshold (same as TUI)
- Add RPC test for manual compaction via compact command
- Auto-compaction emits compaction event with auto: true flag
- Add /compact command for manual context compaction with optional custom instructions
- Add /autocompact command to toggle automatic compaction
- Auto-trigger compaction when context usage exceeds threshold (contextWindow - reserveTokens)
- Add CompactionComponent for TUI display with collapsed/expanded states
- Add compaction events to HTML export with collapsible summary
- Refactor export-html.ts to eliminate duplication between session and streaming formats
- Use setTimeout to break out of agent event handler for safe async compaction
- Show compaction summary in TUI after compaction completes
fixes#92
- Add CompactionEntry type with firstKeptEntryIndex
- Add loadSessionFromEntries() for compaction-aware loading
- Add compact() function that returns CompactionEntry
- Add token calculation and cut point detection
- Add tests with real session fixture and LLM integration
Still TODO: settings, /compact and /autocompact commands, auto-trigger in TUI, /branch rework
pi-ai:
- Fixed usage.input to exclude cached tokens for OpenAI providers
- Previously input included cached tokens, causing double-counting
- Now input + output + cacheRead + cacheWrite correctly gives total context
coding-agent:
- Session header now includes branchedFrom field for branched sessions
- Updated compaction.md with refined implementation plan
- Updated session.md with branchedFrom documentation
- Add piConfig to package.json for app name and config directory
- Consolidate paths.ts into config.ts with clearer naming
- Fix Bun binary detection (changed from %7EBUN to $bunfs)
- Update all hardcoded paths to use config.ts exports
- getThemesDir() for built-in themes, getCustomThemesDir() for user themes
- Strip carriage return characters from bash tool output to fix background padding on Windows
- Add hidden /debug command to write rendered lines to debug log for TUI debugging
- Document /debug command in README.md development section
- Add build:binary script for Bun compilation
- Add paths.ts for cross-platform asset resolution (npm/bun/tsx)
- Add GitHub Actions workflow for automated binary releases
- Update README with installation options
Based on #89 by @steipete
Mario Zechner