agent/pi
2
0
Fork 0
mirror of https://github.com/earendil-works/pi.git synced 2026-06-18 15:54:04 +08:00
Code Issues Packages Projects Releases Wiki Activity
3,794 Commits 32 Branches 288 Tags
91154d9757c1da8a7b2ba5be4bf025fe58d53114
Commit Graph
3452 Commits
Author SHA1 Message Date
Mario Zechner 91154d9757 fix: update stale CI test expectations 2026-04-24 20:22:16 +02:00
Aliou DialloandGitHub 39a9784d2a fix(coding-agent): preserve models when re-registering provider with overrides only (#3651) 
Previously, registerProvider() overwrote the stored config for a provider,
so a models registration followed by a baseUrl/headers-only override lost
the models. After refresh(), only the override-only config was replayed,
causing extension-provided models to disappear.

Added upsertRegisteredProvider() that merges defined fields into the
existing stored config instead of replacing it. Fields absent from the
incoming config are preserved from the stored config.

Also adds regression tests for dynamic provider override persistence
across refresh.
 2026-04-24 20:15:35 +02:00
Mario Zechner 9b103e5e41 fix(ai): replay DeepSeek V4 reasoning content 
closes #3668
 2026-04-24 19:32:51 +02:00
Mario Zechner b4e93feb8c fix(ai): preserve DeepSeek V4 Pro xhigh reasoning 
closes #3662
 2026-04-24 18:55:12 +02:00
Armin Ronacher 35eae2578c doc: Add documentation index 2026-04-24 18:09:27 +02:00
Armin Ronacher 39db145495 fix(coding-agent): show models.json auth in login selector 2026-04-24 17:28:40 +02:00
HQideaGitHub槐聚Claude Opus 4.7Mario Zechner
 		3e0ee69b5e fix(ai): omit tools field instead of sending empty array (#3650) 
DashScope / Aliyun Qwen (OpenAI-compatible) rejects `tools: []`
with HTTP 400 `"[] is too short - 'tools'"`. Five providers used
a truthy check (`if (context.tools)`) that treated an empty array
as "send tools", so `pi --no-tools` produced `tools: []` in the
request body. Matching the Google provider's pattern, we now
guard on `context.tools.length > 0`:

- openai-completions.ts
- openai-responses.ts
- openai-codex-responses.ts
- azure-openai-responses.ts
- anthropic.ts

The openai-completions fallback that emits `tools: []` when the
conversation has tool history (required by LiteLLM / Anthropic
proxies) is preserved via the existing `else if (hasToolHistory)`
branch.

closes #3649

Co-authored-by: 槐聚 <huaiju@zbyte-inc.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Co-authored-by: Mario Zechner <badlogicgames@gmail.com>
 2026-04-24 14:24:31 +02:00
Mario Zechner 6160892626 Add [Unreleased] section for next cycle 2026-04-24 14:20:18 +02:00
Mario Zechner 48aa882b5a Release v0.70.2 2026-04-24 14:19:20 +02:00
Mario Zechner c1b62ae7fe fix(ai,coding-agent): omit undefined provider request options closes #3627 2026-04-24 14:18:35 +02:00
Mario Zechner 90d8051636 Add [Unreleased] section for next cycle 2026-04-24 14:09:27 +02:00
Mario Zechner 6b443c685f Release v0.70.1 2026-04-24 14:08:25 +02:00
Mario Zechner b121235379 chore: audit unreleased changelogs 2026-04-24 14:07:35 +02:00
Mario Zechner 7bd9f6500d fix(coding-agent): avoid cmd.exe for git installs on windows closes #3642 2026-04-24 13:56:10 +02:00
Mario Zechner c06750410a fix(ai,coding-agent): expose provider timeout/retry controls closes #3627 2026-04-24 13:33:00 +02:00
GeorgiGitHubGeorgi Chochev
 		d06db09a53 fix(tui): decode CSI-u Ctrl+letter inside bracketed paste, fixes #3599 (#3623) 
Co-authored-by: Georgi Chochev <georgi.chochev@siteground.com>
 2026-04-24 13:30:49 +02:00
Mario Zechner 6e6c9b4d28 docs(coding-agent): document DeepSeek provider support 
closes #3636
 2026-04-24 12:55:58 +02:00
Mario Zechner 1e33492525 fix(coding-agent): harden clipboard copy 
closes #3639
 2026-04-24 12:55:58 +02:00
Mario Zechner ebbc67df5d docs(coding-agent): fix extension flag lookup docs 
closes #3614
 2026-04-24 11:41:51 +02:00
Mario Zechner c96c2fcd1e fix(ai): correct gpt-5.5 context metadata 2026-04-24 10:42:44 +02:00
Mario Zechner 0fce64d425 Add [Unreleased] section for next cycle 2026-04-24 01:01:39 +02:00
Mario Zechner 612be54c28 Release v0.70.0 2026-04-24 01:00:41 +02:00
Mario Zechner 55fbd9b286 chore(coding-agent): audit and update changelog entries since v0.69.0 2026-04-24 00:59:08 +02:00
Mario Zechner f2f03616ab fix(coding-agent): update provider auth guidance and defaults 2026-04-24 00:55:27 +02:00
Mario Zechner 2926f7e887 chore: update generated models 2026-04-24 00:55:27 +02:00
Mario Zechner bffb760f4f fix(coding-agent): improve auth setup warnings 2026-04-24 00:55:27 +02:00
Armin Ronacher 27c05b7faa feat(ui): Lower /scoped-models height to same as /models 2026-04-24 00:53:11 +02:00
Armin Ronacher d4d3c2fd39 feat(login): More consistent styling 2026-04-24 00:30:31 +02:00
Armin Ronacher 40edf7ee06 fix: /model cancellation 2026-04-24 00:16:10 +02:00
Mario Zechner d7fb9aaf1e docs(coding-agent): audit unreleased changelog 2026-04-24 00:12:07 +02:00
Mario Zechner 6bb4a55524 fix(coding-agent): preserve final TUI frame on quit 2026-04-24 00:11:09 +02:00
Mario Zechner 65a6472bdd fix(ai): forward google vertex base url 
closes #3619
 2026-04-24 00:05:43 +02:00
Mario Zechner 95f4921250 fix(ai): repair codex stream test merge conflict 2026-04-23 23:51:20 +02:00
Markus YlisiurunenGitHubMario Zechner
 		ba8f1c228a Align GPT-5.5 Codex capability handling (#3618) 
* fix(ai): align gpt-5.5 codex capabilities

* chore(ai): update generated models

* fix(ai): apply gpt-5.5 priority pricing to api

---------

Co-authored-by: Mario Zechner <badlogicgames@gmail.com>
 2026-04-23 23:49:33 +02:00
Mario Zechner 86ba08a2e3 fix(coding-agent): preserve in-memory settings on reload 
closes #3616
 2026-04-23 23:47:49 +02:00
Mario Zechner 4cd4cfd98e fix(ai): support long cache retention compat 
closes #3543
 2026-04-23 23:43:34 +02:00
Mario Zechner 1312346199 fix(ai): expand Copilot eager streaming compat 
closes #3575
 2026-04-23 23:43:00 +02:00
Mario Zechner 76bc605ad1 fix(tui): keep terminal progress alive 
closes #3610
 2026-04-23 23:39:06 +02:00
Mario Zechner 19f2478599 fix(coding-agent): show shortcut conflict diagnostics on startup 
closes #3617
 2026-04-23 23:31:52 +02:00
Mario Zechner c681d35d76 fix(ai): avoid double-counting reasoning tokens 
closes #3581
 2026-04-23 23:28:16 +02:00
Mario Zechner ffa0f31239 fix(ai): support Anthropic eager tool streaming compat 
closes #3575
 2026-04-23 23:12:45 +02:00
Mario Zechner 6af10c9c7f fix(ai): make OpenAI Responses session_id header optional closes #3579 2026-04-23 23:12:24 +02:00
Mario Zechner bf4aa3a601 fix(ai): support xhigh for Codex GPT-5.5 2026-04-23 22:49:09 +02:00
Mario Zechner 8700ac1f0e fix(coding-agent): support custom npmCommand for git deps 
closes #3604
 2026-04-23 22:32:32 +02:00
Mario Zechner c091aa732f fix(coding-agent): space first user message after notices 
closes #3613
 2026-04-23 22:16:40 +02:00
Armin Ronacher 00cb947166 fix(ai): preserve env api key contract 2026-04-23 22:11:46 +02:00
Mario Zechner f0cf8a59d2 fix(coding-agent): handle stale extension contexts 
fixes #3606
 2026-04-23 22:07:13 +02:00
Armin Ronacher 0cb8ff3843 fix(ai): Restore original api contract for getEnvApiKey 2026-04-23 22:04:42 +02:00
Mario Zechner 2394f3a987 fix(coding-agent): retry Bedrock http2 no-response errors closes #3594 2026-04-23 21:36:51 +02:00
Mario Zechner 535fdcea2c test(ai): refresh responses and codex e2e coverage closes #3576 2026-04-23 21:36:16 +02:00