bb3d7d399c
feat(coding-agent): expose session metadata to bash tools ( #6967 )
2026-07-22 20:15:02 +02:00
c55ae2faa5
fix(coding-agent): display path of sibling dependent extensions ( #6964 )
...
fixes : #6619
2026-07-22 16:08:21 +02:00
75e6123aba
fix(coding-agent): speed up external editor launch ( #6903 )
...
* fix(coding-agent): speed up external editor launch
Fixes #6774
* refactor(coding-agent): clarify external editor handling
* refactor(coding-agent): simplify external editor result
* refactor(coding-agent): model external editor lifecycle
* fix(coding-agent): rename external editor test fake
* fix(coding-agent): secure external editor temp files
* fix(coding-agent): simplify external editor temp path
* fix(coding-agent): restore system temp editor files
2026-07-22 15:49:01 +02:00
fe42ba5b38
write tui debug/crash logs into the configured pi agent dir ( #6958 )
...
fixes #6652
2026-07-22 15:48:25 +02:00
b9e5c5d941
fix(agent): restore streamFn extension compatibility
...
Keep streamFn required for typed callers while preserving the legacy runtime fallback for extensions that omit it.\n\nfixes #6915
2026-07-21 18:37:07 +02:00
b142504125
fix(coding-agent): defer catalog refresh until after TUI startup
2026-07-21 18:25:38 +02:00
ceac988635
Merge remote-tracking branch 'origin/main' into fix/issue-6647-retry-summary-requests-2
2026-07-21 17:15:54 +02:00
864b35c462
fix(coding-agent): show llama download progress
2026-07-21 14:25:10 +02:00
162179af5a
add branch/compact retries to agent-harness
2026-07-21 12:11:39 +00:00
8e53e0e49c
compaction & branch summarization follow retry policy
...
fixes #6647
compaction (auto & manual) and branch summarization retry on transient failures.
use the same retry policy from settings.
emit events for the tui to show indication of retries
2026-07-21 10:07:20 +00:00
54fad505b9
fix(coding-agent): prefer newer generated model catalogs
2026-07-21 11:11:08 +02:00
1235c0ec64
fix(agent): decouple agent streams from compat
...
closes #6851
2026-07-20 17:54:36 +02:00
c889eb8809
fix(coding-agent): defer startup model catalog refresh
2026-07-20 17:13:06 +02:00
f8b74a4507
fix: complete extension usage accounting
...
closes #6509
2026-07-20 17:06:22 +02:00
2fd3868401
add usage info to branch summary, compaction and tool result entries ( #6671 )
...
* add usage info to branch summary entries
* add usage to compaction entries
* allow custom tools to report llm usage in tool results
* allow observing and patching usage in tool_result hooks
* agent-harness: save usage in entries
for compaction, branch summaries and tool results
2026-07-20 16:41:43 +02:00
c179395218
feat: get_available_thinking_levels rpc ( #6865 )
2026-07-20 16:29:10 +02:00
bbb91fa8ae
feat(ai): add Qwen Token Plan as built-in provider ( #6858 )
...
Add Alibaba Cloud Model Studio Token Plan subscription service as two
built-in API-key providers: qwen-token-plan (international, Singapore)
and qwen-token-plan-cn (China, Beijing).
Each provider exposes 15 text-generation models (Qwen, DeepSeek, GLM,
Kimi, MiniMax) via the OpenAI-compatible endpoint with DashScope
enable_thinking support. Model metadata is sourced from models.dev;
qwen3.8-max-preview is hardcoded until models.dev includes it.
Also fixes kimi-coding test references (k2p7 -> kimi-for-coding) after
models.dev catalog update picked up by generate-models.
Closes #6850
2026-07-20 13:53:30 +02:00
9e7fce70a3
export missing message and tool execution event types ( #6772 )
...
fixes #6687
2026-07-20 13:09:12 +02:00
94373d815d
feat(ai): add shared contentText utility ( #6840 )
...
Co-authored-by: Armin Ronacher <armin.ronacher@active-4.com >
2026-07-20 01:50:36 +02:00
d2f8dafb0f
fix(ai,agent,coding-agent): share UUIDv7 and use for Codex ( #6834 )
2026-07-20 01:23:47 +02:00
f1c587dde3
fix(coding-agent): avoid duplicate session reads
...
fixes #6793
2026-07-19 20:18:49 +02:00
3da591ab74
feat(coding-agent): add Hugging Face llama search
2026-07-17 17:13:39 +02:00
f1a466b19d
feat(coding-agent): add llama.cpp router integration
2026-07-17 16:25:00 +02:00
64f83c85d9
fix(coding-agent): support all-argument prompt defaults closes #6695
2026-07-17 13:52:32 +02:00
019e4ad687
feat(coding-agent): accept native extension providers
...
Allow extensions to register complete pi-ai Provider objects while preserving models.json composition and resolved provider auth access.
Refs #4823 and #4824 .
2026-07-17 10:58:48 +02:00
8881e17625
fix(ai): show implied Kimi Coding subscription costs
2026-07-17 10:01:22 +02:00
35a0d5d620
fix(coding-agent): preserve compaction queue behavior ( #6730 )
2026-07-17 09:09:01 +02:00
a2c5ee33eb
fix(coding-agent): do not highlight read errors ( #6731 )
2026-07-17 09:02:50 +02:00
b97ed20210
fix(coding-agent): explain clone failure for unsaved sessions
2026-07-16 19:06:56 +02:00
a01baaaea7
feat(xai): prefilled OAuth device link, SuperGrok login label, trimmed model list ( #6734 )
...
- Use verification_uri_complete from the xAI device-code response so the
login link opens with the user code prefilled (https-validated)
- Add OAuthAuth.loginLabel and show "Sign in with SuperGrok or X Premium"
in the xAI auth-type selector (threaded through lazyOAuth)
- Remove grok-3, grok-3-fast, grok-4.20 variants and grok-code-fast-1
from the built-in xAI catalog and default to grok-4.5
Co-authored-by: Jaaneek <Jaaneek@users.noreply.github.com >
2026-07-16 18:59:54 +02:00
3524cd4c85
Merge remote-tracking branch 'origin/main' into add-kimi-deferred-tools
2026-07-16 17:30:50 +02:00
6442536b1e
fix(coding-agent): bundle OAuth flows in Bun binaries
2026-07-16 15:36:16 +02:00
f16b4e0cda
add deferred tools support for kimi in openai-completions api
2026-07-16 12:26:56 +00:00
97f9978fa6
feat(coding-agent): add model catalog refresh flag
2026-07-16 12:46:49 +02:00
12545274ea
windows: reset pi terminal title after checking npm packages
...
on windows the npm check changes the terminal title and it doesn't change back
this is a narrow fix for it
fixes #6629
2026-07-15 15:14:55 +00:00
bd9e09db44
feat(coding-agent): expose dynamic provider refresh
2026-07-15 12:56:50 +02:00
45203abfa0
fix(coding-agent): Coalesce adjacent assistant thinking blocks into one section
2026-07-15 12:52:53 +02:00
fab309e955
fix(coding-agent): refresh model catalogs in picker
2026-07-15 12:35:29 +02:00
cd7cad4ee2
feat(coding-agent): merge origin/main into model runtime facade
2026-07-15 12:25:36 +02:00
9993c96907
feat(coding-agent): replace model registry with model runtime
...
Move provider auth and OAuth flows onto pi-ai Models, compose models.json and extension overlays through ModelRuntime, and retain ModelRegistry as an extension compatibility facade.
2026-07-14 17:48:45 +02:00
f4e9ca7466
remove current date from system prompt
...
fixes #6621
2026-07-14 14:58:24 +00:00
adfac437bb
fix(coding-agent): clarify login options
2026-07-14 12:24:50 +02:00
961fa6c142
feat(ai): add Radius gateway support
2026-07-14 11:01:21 +02:00
b084d2fb39
add legacy-peer-deps flag on pi uninstall when using npm ( #6604 )
...
otherwise it would fail when there's conflicting peer dependencies in installed packages
fixes : #6486
2026-07-13 12:50:46 +02:00
7303cbac5d
fix branch summary when using ambient auth ( #6595 )
...
allow null apiKey. uses the same auth flow like compaction
fixes #6324
2026-07-13 10:49:10 +02:00
298665cfb9
fix(ai): support OpenRouter session affinity ( #6496 )
...
Fixes #6366
2026-07-13 09:57:53 +02:00
3d8f74357c
feat(ai): support message-anchored tool loading ( #6474 )
...
This adds cache-friendly dynamic tool loading anchored to tool results. Purely additive active-tool changes are recorded with `addedToolNames`, allowing supported Anthropic and OpenAI Responses models to load tool definitions at the point they become available instead of placing them in the cached prompt prefix.
It retains safe fallback behavior for unsupported models and non-additive changes but it will wipe caches.
2026-07-10 23:52:54 +02:00
d7a48d30a0
fix(coding-agent): fall back to text clipboard paste
2026-07-10 21:46:15 +02:00
3b686ac224
feat(coding-agent): add message copy shortcut
2026-07-10 21:35:53 +02:00
3ea064ea2a
fix: support Bedrock API key login
2026-07-10 19:34:11 +02:00
Armin Ronacher