1754 Commits

• Release v0.50.6
  Mario Zechner · 2026-01-30 22:03:14 +01:00
  f8d43ce02c
• docs: add missing changelog entries for v0.50.5..HEAD

  - tui: add #1095 content shrink fix (by @marckrenn)
  - tui: add isImageLine perf optimization
  - coding-agent: cross-package duplication for #1095, #1099

  Mario Zechner · 2026-01-30 22:02:17 +01:00
  d91cdc5f0c
• feat(coding-agent): add ctx.getSystemPrompt() to extension context

  Adds a method to access the effective system prompt (after any per-turn
  extension modifications) from the extension context.
  
  Implementation:
  - Add systemPrompt getter to AgentSession reading from agent.state.systemPrompt
  - Wire getSystemPrompt through ExtensionContextActions to ExtensionRunner
  - Add getSystemPrompt to interactive-mode's shortcut context
  - Update docs with ctx.getSystemPrompt() section
  - Add system-prompt-header.ts example
  - Add example to docs reference table
  
  Closes #1098

  Mario Zechner · 2026-01-30 17:44:25 +01:00
  99281e5913
• Update providers.md to include opencode provider (#1094)

  add 'opencode' provider configuration to the documentation.

  Emanuel Stadler · 2026-01-30 16:43:32 +01:00
  fd6c6d8772
• Add [Unreleased] section for next cycle
  Mario Zechner · 2026-01-30 11:48:16 +01:00
  0091857f8b
• Release v0.50.5
  Mario Zechner · 2026-01-30 11:47:35 +01:00
  01d521c625
• Add [Unreleased] section for next cycle
  Mario Zechner · 2026-01-30 03:27:09 +01:00
  2cee7e17de
• Release v0.50.4
  Mario Zechner · 2026-01-30 03:26:23 +01:00
  3f5fe42d37
• feat(coding-agent): add daxnuts easter egg
  Mario Zechner · 2026-01-30 03:25:00 +01:00
  0e2f90b734
• docs: add missing changelog entries for commits since v0.50.3

  - ai: Vercel AI Gateway routing support (#1051)
  - tui: Ctrl+B/Ctrl+F keybindings (#1053), character jump navigation (#1074),
    line boundary navigation (#1050), performance optimizations (#1084)
  - coding-agent: set_session_name RPC (#1075), config selector folder names (#1064)
  - Cross-package duplication of tui/ai changes to coding-agent
  - Add New Features section to coding-agent changelog

  Mario Zechner · 2026-01-30 03:25:00 +01:00
  355bc5adc3
• feat(coding-agent): add OSC 52 clipboard support for SSH/mosh sessions

  Fixes #1069

  Mario Zechner · 2026-01-30 03:25:00 +01:00
  9ee58fdea9
• feat(ai): add Vercel AI Gateway routing support (#1051)

  * feat(ai): add Vercel AI Gateway routing support
  
  Add vercelGatewayRouting to OpenAICompletionsCompat, parallel to
  openRouterRouting. When a model targets ai-gateway.vercel.sh and has
  vercelGatewayRouting configured, the openai-completions provider passes
  providerOptions.gateway with only/order in the request body.
  
  Changes:
  - types.ts: VercelGatewayRouting interface + field on OpenAICompletionsCompat
  - openai-completions.ts: buildParams passes providerOptions.gateway,
    detectCompat/getCompat include the new field
  - model-registry.ts: VercelGatewayRoutingSchema for models.json validation
  - test: updated Required<OpenAICompletionsCompat> in test fixture
  
  * docs(coding-agent): add vercelGatewayRouting to custom models documentation

  Ben Vargas · 2026-01-30 01:44:51 +01:00
  e045a9f142
• feat(tui,coding-agent): add ctrl+b/ctrl+f cursor navigation keybindings (#1053)
  Nathan Igo · 2026-01-30 01:43:54 +01:00
  902473797d
• fix(coding-agent): show folder name for extensions in config selector (#1064)
  Umberto B. · 2026-01-30 01:43:01 +01:00
  af4cf5f4ef
• feat(tui): add character jump navigation (Ctrl+], Ctrl+Alt+]) (#1074)

  Add Bash/Readline-style character search:
  
  - Ctrl+] enters forward jump mode, awaits next character, jumps to it
  - Ctrl+Alt+] does the same but searches backward
  - Multi-line search
  - Case-sensitive matching
  - Pressing the hotkey again cancels; control chars cancel and fall
    through

  Sviatoslav Abakumov · 2026-01-30 01:42:14 +01:00
  c5d16fe456
• feat(coding-agent): add set_session_name RPC command (#1075)

  - Add set_session_name command with empty name validation
  - Expose sessionName in get_state response
  - Add setSessionName() to AgentSession and RpcClient
  - Document in docs/rpc.md

  Daniel Nouri · 2026-01-30 01:41:58 +01:00
  e20583aac8
• feat(coding-agent): add "none" option to doubleEscapeAction setting

  Allows disabling double-escape behavior entirely for users who
  accidentally trigger the tree/fork selector.
  
  Fixes #973

  Mario Zechner · 2026-01-30 01:20:14 +01:00
  cb08758696
• docs(coding-agent): add changelog for tool arg defaults
  Mario Zechner · 2026-01-30 01:16:24 +01:00
  0587f045d9
• fix(coding-agent): respect .gitignore/.ignore/.fdignore when scanning package resources

  fixes #1072

  Mario Zechner · 2026-01-30 01:02:56 +01:00
  098f396cf3
• fix(ai): update Antigravity User-Agent to 1.15.8

  fixes #1079

  Mario Zechner · 2026-01-29 23:10:49 +01:00
  52532c7c00
• fix(coding-agent): handle macOS filenames with curly quotes and NFD Unicode

  Fixes #1078
  
  - Add tryNFDVariant() to normalize paths to NFD form
  - Add tryCurlyQuoteVariant() to convert straight quotes to curly quotes
  - Try combined NFD + curly quote variant for French macOS screenshots
  - Add comprehensive tests for path-utils

  Mario Zechner · 2026-01-29 22:58:22 +01:00
  4edb506df8
• Release v0.50.3
  Mario Zechner · 2026-01-29 04:19:02 +01:00
  df667b510a
• docs(coding-agent): update changelog with missing fix entry and new features section

  - Add New Features section for Kimi For Coding provider
  - Add fix entry for resource ordering when resuming sessions

  Mario Zechner · 2026-01-29 04:16:55 +01:00
  b78b21c588
• feat(ai): add Kimi For Coding provider support

  - Add kimi-coding provider using Anthropic Messages API
  - API endpoint: https://api.kimi.com/coding/v1
  - Environment variable: KIMI_API_KEY
  - Models: kimi-k2-thinking (text), k2p5 (text + image)
  - Add context overflow detection pattern for Kimi errors
  - Add tests for all standard test suites

  Mario Zechner · 2026-01-29 04:12:28 +01:00
  87ab5c5c3b
• fix(coding-agent): show resources before messages when resuming session

  Swapped order of initExtensions() and renderInitialMessages() in init()
  so loaded resources (Context, Skills, Prompts, Extensions) appear at the
  top of the chat instead of at the bottom when resuming a session.

  Mario Zechner · 2026-01-29 03:55:44 +01:00
  d1e33599f6
• Add [Unreleased] section for next cycle
  Mario Zechner · 2026-01-29 03:24:32 +01:00
  ded2d6399e
• Release v0.50.2
  Mario Zechner · 2026-01-29 03:23:41 +01:00
  9bb627a524
• docs: add missing changelog entries for v0.50.1..HEAD commits
  Mario Zechner · 2026-01-29 03:22:00 +01:00
  58aa7096dd
• fix: autocompleteMaxVisible setting not persisting

  Missing markModified() call caused the setting to be lost on reload.

  Mario Zechner · 2026-01-29 03:20:28 +01:00
  2ad4825653
• feat: add autocompleteMaxVisible setting for configurable dropdown height
  Colin Mason · 2026-01-29 03:16:52 +01:00
  b212314f45
• docs(coding-agent): add changelog entry for #1039
  Mario Zechner · 2026-01-29 02:52:43 +01:00
  06a7fedda5
• create headerContainer for custom header (#1039)
  Ted · 2026-01-29 02:52:08 +01:00
  d88087334f
• fix: preserve external settings.json edits on reload (#1046)

  Co-authored-by: Mario Zechner <badlogicgames@gmail.com>

  Nico Bailon · 2026-01-29 02:42:23 +01:00
  bac57f81be
• feat(ai): add Hugging Face provider support

  - Add huggingface to KnownProvider type
  - Add HF_TOKEN env var mapping
  - Process huggingface models from models.dev (14 models)
  - Use openai-completions API with compat settings
  - Add tests for all provider test suites
  - Update documentation
  
  fixes #994

  Mario Zechner · 2026-01-29 02:40:14 +01:00
  c808de605a
• feat(ai): add PI_CACHE_RETENTION env var for extended prompt caching

  Adds support for extended cache retention via PI_CACHE_RETENTION=long:
  - Anthropic: 5m -> 1h TTL
  - OpenAI: in-memory -> 24h retention
  
  Only applies to direct API calls (api.anthropic.com, api.openai.com).
  Proxies and other providers are unaffected.
  
  fixes #967

  Mario Zechner · 2026-01-29 02:22:06 +01:00
  1b6a147579
• fix(coding-agent): reset retry counter after each successful LLM response

  Previously, within a single tool-use turn, rate limit retries would
  accumulate across separate LLM calls. For example, if each of 3 tool
  calls hit a 429 and retried once, the counter would show '3/3' and fail
  even though each individual retry succeeded.
  
  Now the counter resets immediately when a successful (non-error)
  assistant message arrives, so each LLM call gets a fresh set of retries.
  
  Fixes #1019

  Mario Zechner · 2026-01-29 01:12:29 +01:00
  4f004adefa
• fix(coding-agent): improve thinking defaults in footer
  Mario Zechner · 2026-01-29 00:33:45 +01:00
  a373dce747
• feat(tui,coding-agent): add shell-style keybindings alt+b, alt+f, ctrl+d

  - Add alt+b/alt+f as alternative bindings for cursorWordLeft/cursorWordRight
  - Add ctrl+d as alternative binding for deleteCharForward
  - Fix ctrl+d in custom editor to perform delete when text is present
    (previously it was always consumed, even with non-empty input)
  
  Closes #1043
  
  Co-authored-by: Jason Ish <ish@unx.ca>

  Mario Zechner · 2026-01-29 00:21:39 +01:00
  81be81328c
• Closes #1041, incorrect .md file name in warning messages.
  Mario Zechner · 2026-01-29 00:16:50 +01:00
  5ef79459dc
• fix(coding-agent): empty array in package filter now disables all resources

  fixes #1044

  Mario Zechner · 2026-01-29 00:12:42 +01:00
  d21cbde87f
• Add get_commands RPC for headless clients (#995)

  * Add get_commands RPC for headless clients
  
  Headless clients like Emacs can now query which commands are available.
  Previously they could only discover file-based prompt templates by
  scanning the filesystem; extension commands and skills were invisible.
  
  The response includes each command's name, description, and source
  (extension, template, or skill). Commands appear in the same order
  as the TUI's autocomplete: extension commands first, then templates,
  then skills.
  
  Built-in TUI commands (/settings, /fork, etc.) are excluded since
  they require the interactive UI. Commands like /compact have dedicated
  RPC equivalents instead.
  
  * Add location and path to get_commands response
  
  Clients can show where commands come from (user/project/path) and
  display file paths in tooltips. The data is already available on
  templates and skills - just exposing it.

  Daniel Nouri · 2026-01-28 02:34:15 +01:00
  0ad189f12a
• fix(coding-agent): hide provider in footer when terminal is narrow (#981)
  Sviatoslav Abakumov · 2026-01-28 02:15:27 +01:00
  c67b582fc4
• Add [Unreleased] section for next cycle
  Mario Zechner · 2026-01-27 00:15:42 +01:00
  2cc2544809
• Release v0.50.1
  Mario Zechner · 2026-01-27 00:14:58 +01:00
  29c59d79ba
• docs(coding-agent): clarify dependency categories
  Mario Zechner · 2026-01-27 00:13:54 +01:00
  bec72c428c
• docs(coding-agent): clarify pi package dependency bundling
  Mario Zechner · 2026-01-27 00:10:58 +01:00
  484b605ca0
• chore: update changelogs
  Mario Zechner · 2026-01-27 00:08:06 +01:00
  6159eb4be8
• fix(coding-agent): prune empty git dirs on remove
  Mario Zechner · 2026-01-26 23:58:14 +01:00
  ee7d03cf52
• fix(coding-agent): bind extension UI context on startup
  Mario Zechner · 2026-01-26 23:00:48 +01:00
  5dbeadae05
• fix(coding-agent): sync agent state after newSession setup callback

  Move setup callback handling from interactive/rpc modes into AgentSession.newSession().
  After setup() runs, sync agent state via replaceMessages() so the LLM has context
  and the UI renders the messages properly.
  
  fixes #968

  Mario Zechner · 2026-01-26 22:25:09 +01:00
  f86e3c3103