070ef9c771
Release v0.40.0
2026-01-09 00:43:32 +01:00
36fbce85c2
Add [Unreleased] section for next cycle
2026-01-09 00:34:20 +01:00
39bfbb125c
Release v0.39.1
2026-01-09 00:33:39 +01:00
307f9b4169
Add [Unreleased] section for next cycle
2026-01-09 00:10:41 +01:00
c25f1eab64
Release v0.39.0
2026-01-09 00:10:00 +01:00
16e142ef7d
fix(ai): remove <thinking> tag wrapping, convert to plain text on cross-model handoff
...
- Remove <thinking> tag generation from google-shared.ts, transorm-messages.ts, openai-completions.ts
- Thinking blocks now convert to plain text when switching models (prevents models mimicking tags)
- Skip empty thinking blocks to avoid API errors
- Keep thinking blocks only when same provider AND same model
fixes #561
2026-01-08 21:19:16 +01:00
aa89080ea0
fix(ai): add bridge prompt to override Antigravity behavior with Pi defaults
2026-01-08 20:20:24 +01:00
aa91864715
chore: add changelog for #571 , init theme with watcher in interactive mode
2026-01-08 20:10:39 +01:00
ba46a52415
docs: add changelog entries for #568
2026-01-08 18:54:21 +01:00
cfa63c255d
Add [Unreleased] section for next cycle
2026-01-08 04:53:51 +01:00
4b084d0f9d
Release v0.38.0
2026-01-08 04:52:55 +01:00
946efe4b45
fix(ai): detect context_length_exceeded overflow
2026-01-08 03:16:59 +01:00
d311978dfd
Merge branch 'main' into feat/custom-thinking-budgets
2026-01-08 00:39:11 +03:00
39fa25eb67
fix(ai): clean up openai-codex models and token limits
...
- Remove model aliases (gpt-5, gpt-5-mini, gpt-5-nano, codex-mini-latest, gpt-5-codex, gpt-5.1-codex, gpt-5.1-chat-latest)
- Fix context window from 400k to 272k tokens to match Codex CLI defaults
- Keep maxTokens at 128k (original value)
- Simplify reasoning effort clamping
closes #536
2026-01-07 20:39:46 +01:00
d893ba7f20
fix(ai): clean up openai-codex models and token limits
2026-01-07 23:09:20 +08:00
acec56c6b5
feat: add customizable thinking budgets for token-based providers
...
Fixes #521
2026-01-07 15:17:18 +03:00
0f27eae77e
feat: add thinkingBudgets option to customize token budgets
2026-01-07 15:13:26 +03:00
fcb3b4aa72
Add [Unreleased] section for next cycle
2026-01-07 01:33:34 +01:00
5514d4ffdb
Release v0.37.8
2026-01-07 01:32:53 +01:00
7508a51599
Add [Unreleased] section for next cycle
2026-01-07 01:02:34 +01:00
a4f2bb5c27
Release v0.37.7
2026-01-07 01:01:55 +01:00
9330f5a10a
Add [Unreleased] section for next cycle
2026-01-07 00:43:25 +01:00
befbeb30e4
Release v0.37.6
2026-01-07 00:42:45 +01:00
f054150ea3
Add changelog entries for bridge prompt export ( #510 )
2026-01-07 00:11:21 +01:00
a00322175f
Add [Unreleased] section for next cycle
2026-01-06 23:25:53 +01:00
fe95983dcf
Release v0.37.5
2026-01-06 23:25:12 +01:00
f87fb0a38a
Add [Unreleased] section for next cycle
2026-01-06 21:29:03 +01:00
b08a6b22ce
Release v0.37.4
2026-01-06 21:28:24 +01:00
a236e62025
Add [Unreleased] section for next cycle
2026-01-06 13:44:28 +01:00
86fea6379c
Release v0.37.3
2026-01-06 13:43:47 +01:00
edb0da9611
feat(ai,agent,coding-agent): add sessionId for provider session-based caching
...
- Add sessionId to StreamOptions for providers that support session-based caching
- OpenAI Codex provider uses sessionId for prompt_cache_key and routing headers
- Agent class now accepts and forwards sessionId to stream functions
- coding-agent passes session ID from SessionManager and updates on session changes
- Update ai package README with table of contents, OpenAI Codex OAuth docs, and env vars table
- Increase Codex instructions cache TTL from 15 minutes to 24 hours
- Add tests for sessionId forwarding in ai and agent packages
2026-01-06 11:08:42 +01:00
9063a71fe6
Add [Unreleased] section for next cycle
2026-01-06 00:54:00 +01:00
6f4d16d76f
Release v0.37.2
2026-01-06 00:53:18 +01:00
aad93a1ec8
docs: add changelog entry for #484
2026-01-06 00:51:17 +01:00
2da853e9d6
Add [Unreleased] section for next cycle
2026-01-06 00:32:56 +01:00
a20112f45f
Release v0.37.1
2026-01-06 00:32:09 +01:00
0c712659b3
Add [Unreleased] section for next cycle
2026-01-06 00:07:57 +01:00
d3d53392e8
Release v0.37.0
2026-01-06 00:07:15 +01:00
0b9e3ada0c
fix: clean up Codex thinking level handling
...
- Remove per-thinking-level model variants (gpt-5.2-codex-high, etc.)
- Remove thinkingLevels from Model type
- Provider clamps reasoning effort internally
- Omit reasoning field when thinking is off
fixes #472
2026-01-05 21:58:26 +01:00
168d114750
Add changelog entries for headless OAuth and truecolor detection
2026-01-05 20:16:59 +01:00
74d5f5523e
Add [Unreleased] section for next cycle
2026-01-05 06:17:05 +01:00
435cd5d61f
Release v0.36.0
2026-01-05 06:16:25 +01:00
bb50738f7e
fix(ai): append system prompt to codex bridge message instead of converting to input
...
Previously the system prompt was converted to an input message in convertMessages,
then stripped out by filterPiSystemPrompts. Now the system prompt is passed directly
to transformRequestBody and appended after CODEX_PI_BRIDGE in the bridge message.
2026-01-05 06:03:07 +01:00
eede2e1644
Merge origin/main into feature/detect-vertex-ai
2026-01-05 04:35:17 +01:00
32f032b6d1
Add [Unreleased] section for next cycle
2026-01-05 03:42:07 +01:00
78d0b88f7d
Release v0.35.0
2026-01-05 03:41:24 +01:00
63093bf7e4
fix(ai): cache vertex adc credentials file check
2026-01-04 17:16:40 -08:00
a3f30e085a
fix(ai): add vertex ai dummy value for configured credentials
2026-01-04 12:59:19 -08:00
12805f61bd
Add [Unreleased] section for next cycle
2026-01-04 21:06:48 +01:00
71c978e2f9
Release v0.34.2
2026-01-04 21:06:08 +01:00
Mario Zechner