Commit graph

364 commits

Author SHA1 Message Date
Mario Zechner
fd7f20f968 Remove unused BashExecutionMessage import 2025-12-09 01:00:11 +01:00
Mario Zechner
376b858d2b Remove unnecessary type cast in interactive-mode.ts 2025-12-09 00:59:25 +01:00
Mario Zechner
6adfb64279 Fix path resolution in config.ts after file reorganization 2025-12-09 00:54:51 +01:00
Mario Zechner
83a6c26969 Reorganize file structure: core/, utils/, modes/interactive/components/, modes/interactive/theme/ 2025-12-09 00:51:33 +01:00
Mario Zechner
00982705f2 WP16: Update main-new.ts to use InteractiveMode 2025-12-09 00:36:16 +01:00
Mario Zechner
0020de8518 WP15: Create InteractiveMode using AgentSession 2025-12-09 00:34:44 +01:00
Mario Zechner
e9f6de7cb8 WP14: Create main-new.ts and cli-new.ts using AgentSession and new modes 2025-12-09 00:21:18 +01:00
Mario Zechner
7890fb836a WP13: Create modes/index.ts barrel export 2025-12-09 00:16:46 +01:00
Mario Zechner
e7c71e7ee3 WP12: Create rpc-mode.ts 2025-12-09 00:16:17 +01:00
Mario Zechner
c0996a1078 WP11: Create print-mode.ts 2025-12-09 00:15:37 +01:00
Mario Zechner
934c2bc5d3 WP9+WP10: Add AgentSession session management and utility methods 2025-12-09 00:14:47 +01:00
Mario Zechner
94ff0b0962 WP8: Add AgentSession bash execution 2025-12-09 00:13:10 +01:00
Mario Zechner
8d6d2dd72b WP7: Add AgentSession compaction (manual + auto), fix listener preservation 2025-12-09 00:12:07 +01:00
Mario Zechner
0119d7610b WP5+WP6: Add AgentSession model, thinking level, and queue mode management 2025-12-09 00:08:36 +01:00
Mario Zechner
d08e1e53e9 WP4: Add AgentSession prompting methods (prompt, queue, abort, reset) 2025-12-09 00:07:24 +01:00
Mario Zechner
eba196f4ac WP3: Add AgentSession event subscription with session persistence 2025-12-09 00:05:54 +01:00
Mario Zechner
29d96ab25a WP2: Create AgentSession basic structure + update plan for keep-old-code strategy 2025-12-09 00:05:06 +01:00
Mario Zechner
3f305502cd WP1: Create bash-executor.ts with unified bash execution 2025-12-09 00:01:36 +01:00
Mario Zechner
cceb44410e Add queue mode to AgentSession API, clarify setters vs getters 2025-12-08 23:57:56 +01:00
Mario Zechner
4197b0ceea Fix: use AppMessage[] instead of Message[] in AgentSession 2025-12-08 23:56:08 +01:00
Mario Zechner
1507f8b7a3 Add refactoring plan for coding-agent 2025-12-08 23:54:41 +01:00
Markus Ylisiurunen
d56a0463df
use only model id for fuzzy filtering in model selector (#151) 2025-12-08 23:35:31 +01:00
Mario Zechner
33a2bcf203 Release v0.14.2 2025-12-08 23:29:05 +01:00
Mario Zechner
ad42ebf5f5 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.
2025-12-08 23:26:58 +01:00
Mario Zechner
a054fecd11 Release v0.14.1 2025-12-08 22:59:46 +01:00
Mario Zechner
0bc8d79216 Release v0.14.0 2025-12-08 22:51:29 +01:00
Mario Zechner
d5200b4f1c Add green borders around bash execution component 2025-12-08 22:45:05 +01:00
Mario Zechner
bd0d0676d4 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>
2025-12-08 22:40:32 +01:00
Mario Zechner
1608da8770 Merge main into bash-mode 2025-12-08 21:39:01 +01:00
Mario Zechner
00370cab39 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
2025-12-08 21:12:54 +01:00
Mario Zechner
87a1a9ded4 Add OpenAICompat for openai-completions provider quirks
Fixes #133
2025-12-08 19:02:03 +01:00
Nico Bailon
2e3ff4a15a
Fix truncation test assertions to match new message format (#136) 2025-12-07 12:07:15 +01:00
Mario Zechner
a0bbc29201 Release v0.13.2 2025-12-07 01:25:17 +01:00
Mario Zechner
306f9cc660 Add changelog entry for tool output truncation (#134) 2025-12-07 01:24:41 +01:00
Mario Zechner
5a549cc7da 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.
2025-12-07 01:14:57 +01:00
Mario Zechner
b813a8b92b 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)
2025-12-07 01:11:31 +01:00
Mario Zechner
de77cd1419 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
2025-12-07 00:03:16 +01:00
Mario Zechner
95eadb9ed7 Release v0.13.1 2025-12-06 23:12:36 +01:00
Mario Zechner
ecdbd88f5d Release v0.13.0 2025-12-06 22:48:39 +01:00
Mario Zechner
86e5a70ec4 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
2025-12-06 22:46:02 +01:00
badlogic
52f1a8cb31 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
2025-12-06 22:42:47 +01:00
Mario Zechner
301c6ba11f Release v0.12.15 2025-12-06 21:29:03 +01:00
Mario Zechner
d7f84469a7 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.
2025-12-06 21:24:26 +01:00
Mario Zechner
10a1e1ef90 docs: add under-compaction analysis
Documents context window overflow scenarios, how OpenCode and Codex
handle them, and what fixes are needed.

Related to #128
2025-12-06 14:53:05 +01:00
Mario Zechner
4a972fbe6c Release v0.12.14 2025-12-06 00:46:57 +01:00
Mario Zechner
05849258b5 Release v0.12.13 2025-12-05 23:38:32 +01:00
Mario Zechner
bd3f5779d3 Add changelog entry for parallel version check 2025-12-05 23:36:30 +01:00
Mario Zechner
7352072bc2 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.
2025-12-05 23:36:06 +01:00
Mario Zechner
c423734e36 Add [Unreleased] section 2025-12-05 23:01:46 +01:00
Mario Zechner
ddf09720cc Release v0.12.12 2025-12-05 23:00:45 +01:00