697 Commits

This Branch

This Branch

All Branches

• Release v0.14.2
  Mario Zechner · 2025-12-08 23:29:05 +01:00
  v0.14.2 33a2bcf203
• Fix crash when bash mode outputs binary data

  Sanitize shell output by removing Unicode Format characters and lone
  surrogates that crash string-width. This fixes crashes when running
  commands like curl that download binary files.

  Mario Zechner · 2025-12-08 23:26:58 +01:00
  ad42ebf5f5
• Release v0.14.1
  Mario Zechner · 2025-12-08 22:59:46 +01:00
  v0.14.1 a054fecd11
• Fix tsgo type issues: update tsgo, fix ReasoningEffort import, remove broken enum-test
  Mario Zechner · 2025-12-08 22:59:13 +01:00
  238c5d34e4
• Release v0.14.0
  Mario Zechner · 2025-12-08 22:51:29 +01:00
  v0.14.0 0bc8d79216
• Merge branch 'bash-mode'
  Mario Zechner · 2025-12-08 22:49:28 +01:00
  21800035cd
• Add green borders around bash execution component
  Mario Zechner · 2025-12-08 22:45:05 +01:00
  d5200b4f1c
• Add bash mode for executing shell commands

  - Add ! prefix in TUI editor to execute shell commands directly
  - Output streams in real-time and is added to LLM context
  - Supports multiline commands, cancellation (Escape), truncation
  - Preview mode shows last 20 lines, Ctrl+O expands full output
  - Commands persist in session history as bashExecution messages
  - Add bash command to RPC mode via {type:'bash',command:'...'}
  - Add RPC tests for bash command execution and context inclusion
  - Update docs: rpc.md, session.md, README.md, CHANGELOG.md
  
  Closes #112
  
  Co-authored-by: Markus Ylisiurunen <markus.ylisiurunen@gmail.com>

  Mario Zechner · 2025-12-08 22:40:32 +01:00
  bd0d0676d4
• Merge main into bash-mode
  Mario Zechner · 2025-12-08 21:39:01 +01:00
  1608da8770
• fix: include empty tools param when conversation has tool history (#150)

  Anthropic (via LiteLLM/proxy) requires the `tools` parameter to be
  present when messages include tool_calls or tool role messages,
  even if no tools are currently being provided.
  
  This adds a `hasToolHistory()` helper to detect if the conversation
  contains tool calls or tool results, and ensures `tools: []` is
  included in the request params when needed.
  
  Fixes #149
  
  🤖 Generated with [Claude Code](https://claude.com/claude-code)
  
  Co-authored-by: Ubuntu <ubuntu@ip-172-31-50-87.us-west-2.compute.internal>
  Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>

  Tiago Freitas · 2025-12-08 21:30:43 +01:00
  8f67e00160
• Add xhigh thinking level for OpenAI codex-max models

  - Add 'xhigh' to ThinkingLevel type in ai and agent packages
  - Map xhigh to reasoning_effort: 'max' for OpenAI providers
  - Add thinkingXhigh color token to theme schema and built-in themes
  - Show xhigh option only when using codex-max models
  - Update CHANGELOG for both ai and coding-agent packages
  
  closes #143

  Mario Zechner · 2025-12-08 21:12:54 +01:00
  00370cab39
• Add OpenAICompat for openai-completions provider quirks

  Fixes #133

  Mario Zechner · 2025-12-08 19:02:03 +01:00
  87a1a9ded4
• Merge PR #137: Remove provider-level tool validation
  Mario Zechner · 2025-12-08 18:06:50 +01:00
  e34e0c503f
• Remove provider-level tool validation, add validateToolCall helper
  Mario Zechner · 2025-12-08 18:04:33 +01:00
  8bec289dc6
• add option to skip provider tool call validation
  Markus Ylisiurunen · 2025-12-07 17:24:06 +02:00
  0196308266
• Fix truncation test assertions to match new message format (#136)
  Nico Bailon · 2025-12-07 12:07:15 +01:00
  2e3ff4a15a
• Release v0.13.2
  Mario Zechner · 2025-12-07 01:25:17 +01:00
  v0.13.2 a0bbc29201
• Add changelog entry for tool output truncation (#134)
  Mario Zechner · 2025-12-07 01:24:41 +01:00
  306f9cc660
• Restore TUI warning notices for truncated tool output

  Warnings now shown at bottom of tool execution (outside collapsed area)
  so users can see truncation occurred even when content is collapsed.

  Mario Zechner · 2025-12-07 01:14:57 +01:00
  5a549cc7da
• Implement tool result truncation with actionable notices (#134)

  - 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)

  Mario Zechner · 2025-12-07 01:11:31 +01:00
  b813a8b92b
• Add tool output truncation with line/byte limits

  - 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

  Mario Zechner · 2025-12-07 00:03:16 +01:00
  de77cd1419
• Release v0.13.1
  Mario Zechner · 2025-12-06 23:12:36 +01:00
  v0.13.1 95eadb9ed7
• Add [Unreleased] section to CHANGELOG
  Mario Zechner · 2025-12-06 22:49:45 +01:00
  2641424bfa
• Release v0.13.0
  Mario Zechner · 2025-12-06 22:48:39 +01:00
  v0.13.0 ecdbd88f5d
• Add totalTokens field to Usage type

  - 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

  Mario Zechner · 2025-12-06 22:46:02 +01:00
  86e5a70ec4
• Fix Windows binary detection for Bun compiled executables

  - 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

  badlogic · 2025-12-06 22:42:47 +01:00
  52f1a8cb31
• Release v0.12.15
  Mario Zechner · 2025-12-06 21:29:03 +01:00
  v0.12.15 301c6ba11f
• Fix editor crash with wide characters (emojis, CJK)

  Editor text wrapping now uses grapheme-aware width calculation instead
  of string length. Fixes crash when pasting text containing emojis like
  ✅ or CJK characters that are 2 terminal columns wide.

  Mario Zechner · 2025-12-06 21:24:26 +01:00
  d7f84469a7
• Add context overflow detection utilities

  Extract overflow detection logic into reusable utilities:
  - isContextOverflowError() to detect overflow from error messages
  - isContextOverflowFromUsage() to detect overflow from token usage
  - Patterns for Anthropic, OpenAI, Google, xAI, Groq, OpenRouter, llama.cpp, LM Studio
  
  Fixes #129

  Mario Zechner · 2025-12-06 21:24:15 +01:00
  a325c1c7d1
• docs: add under-compaction analysis

  Documents context window overflow scenarios, how OpenCode and Codex
  handle them, and what fixes are needed.
  
  Related to #128

  Mario Zechner · 2025-12-06 14:53:05 +01:00
  10a1e1ef90
• docs: fix mini-lit links (#123)
  Peter Steinberger · 2025-12-06 09:56:31 +01:00
  fa77ef8b6a
• Fix wrap-ansi test to use node:test instead of vitest
  Mario Zechner · 2025-12-06 00:48:46 +01:00
  ee83284dcf
• Release v0.12.14
  Mario Zechner · 2025-12-06 00:46:57 +01:00
  v0.12.14 4a972fbe6c
• Release v0.12.13
  Mario Zechner · 2025-12-05 23:38:32 +01:00
  v0.12.13 05849258b5
• Add changelog entry for parallel version check
  Mario Zechner · 2025-12-05 23:36:30 +01:00
  bd3f5779d3
• Run version check in parallel with TUI startup

  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.

  Mario Zechner · 2025-12-05 23:36:06 +01:00
  7352072bc2
• Add [Unreleased] section
  Mario Zechner · 2025-12-05 23:01:46 +01:00
  c423734e36
• Release v0.12.12
  Mario Zechner · 2025-12-05 23:00:45 +01:00
  v0.12.12 ddf09720cc
• Merge branch 'feat/resume-slash-command'
  Mario Zechner · 2025-12-05 22:54:03 +01:00
  94d6c6704b
• feat(coding-agent): add /resume command to switch sessions mid-conversation

  - Opens interactive session selector
  - Properly aborts in-flight agent turns before switching
  - Restores model and thinking level from resumed session
  - Clears UI state (queued messages, pending tools, etc.)
  
  closes #117

  Mario Zechner · 2025-12-05 22:53:51 +01:00
  4bd52cf3ed
• Merge feat/scroll-previous-prompts with history load on resume
  Mario Zechner · 2025-12-05 22:02:16 +01:00
  30dd31d787
• feat(tui): add prompt history navigation with Up/Down arrows

  - 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

  Mario Zechner · 2025-12-05 22:00:23 +01:00
  3a5185c5fd
• fix: escape codes not properly wrapped during line wrapping

  - ANSI codes now attach to next visible content, not trailing whitespace
  - Rewrote AnsiCodeTracker to track individual attributes
  - Line-end resets only turn off underline, preserving background colors
  - Added vitest config to exclude node:test files
  
  fixes #109

  Mario Zechner · 2025-12-05 21:49:44 +01:00
  6ec1391ebb
• Implement fuzzy search for model/session selector and improve Input multi-key sequence handling (#122)

  * implement fuzzy search and filtering for tui selectors
  
  * update changelog and readme
  
  * add correct pr to changelog

  Markus Ylisiurunen · 2025-12-05 20:33:04 +01:00
  ff047e5ee1
• feat(tui): add prompt history navigation with Up/Down arrows

  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.

  Nico Bailon · 2025-12-05 07:57:42 -08:00
  c550ed2bca
• docs: add changelog entry for footer display changes
  Mario Zechner · 2025-12-05 12:22:30 +01:00
  ca39e899f4
• fix: make footer token counts more compact

  - Use M suffix for millions (e.g., 10.2M instead of 10184k)
  - Change '% of' to '%/' for context display

  Mario Zechner · 2025-12-05 12:21:44 +01:00
  5c388c0a77
• Add [Unreleased] section
  Mario Zechner · 2025-12-05 12:20:35 +01:00
  6a29b2df3f
• Release v0.12.11
  Mario Zechner · 2025-12-05 12:19:37 +01:00
  v0.12.11 b193560ab0
• fix: footer overflow crash on narrow terminals

  Footer stats line now truncates gracefully when terminal width is too
  narrow to fit all stats, instead of overflowing and crashing the TUI.

  Mario Zechner · 2025-12-05 12:17:56 +01:00
  ef333af3d1