mortal/page-agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,103 Commits 1 Branch 96 Tags
4122735bbae02353752e0dbb2471e9620d0d7f52
Commit Graph

58 Commits

Author SHA1 Message Date
Simon 1cab66b898 feat(llms): update model list and patches (#611) 
* fix(llms): gpt patch

* feat(llms): patch `hunyuan` models

* feat(llms): update model list
 2026-07-10 17:08:14 +08:00
Simon d5c994708c chore(llms): gate live model tests behind test:live (#610) 
* chore(llms): gate live model tests behind test:live

Keep *.live.test.ts out of npm test so local CI stays fast; run them
explicitly via npm run test:live when checking provider compatibility.

* chore: skill gh permission
 2026-07-10 15:48:52 +08:00
Simon 9a75ead6ad fix: do not enable OpenRouter reasoning by default 2026-07-03 20:30:57 +08:00
Simon 10276c6e67 fix: new URL should always be in a try block 2026-07-03 20:14:11 +08:00
Simon 5aeb67f6d6 feat(llms): update model list; update patch 2026-07-03 03:04:43 +08:00
Simon 556bc7448a chore(llms): add real-api test for all supported models 2026-07-03 02:29:06 +08:00
Simon 04b84b6e8a fix(llms): model compatibility problem 2026-07-03 02:26:01 +08:00
Simon a47e12c0e0 fix: model patch for GPT and Claude 2026-07-02 22:11:28 +08:00
Simon 8ad719744a feat: rewrite llm config and model patch 2026-07-02 21:31:44 +08:00
Simon 2b679d85cb fix(llms): omitting both reasoning_effort and temperature for chat-latest 2026-06-26 00:12:34 +08:00
artboy 57f64347ac fix(llms): skip reasoning_effort patch for *-chat-latest models 
normalizeModelName reduces gpt-5.2-chat-latest to gpt-52-chat-latest,
which matches the gpt-52 prefix and gets reasoning_effort='none'.
But chat-latest models don't support reasoning_effort, causing
HTTP 400.

Fix: check for 'chat-latest' before applying reasoning_effort patches.

Fixes #563
 2026-06-23 14:26:45 +08:00
Simon 2f343bcc56 feat(llms): add Claude Opus 4.8 support 
Strip temperature for claude-opus-4-8, add tests, list model on website.
 2026-06-08 22:23:56 +08:00
Simon 68ae73d4e6 chore(llms): add vitest unit tests for llms package 2026-06-04 20:27:33 +08:00
Simon 9891e02215 fix(llms): handle AbortError during .json() 2026-06-03 23:31:01 +08:00
Simon 823195ad94 fix(llms): add error handling for invalid JSON responses and schema validation 2026-06-03 23:14:12 +08:00
Simon fd12fb9f1b refactor(llms): split AbortError out of InvokeError 2026-06-03 23:00:33 +08:00
Simon fb8de08c39 fix(llms): fix abort handling and clean up retry logic 2026-06-03 22:44:54 +08:00
Simon 90270fb76c fix(ts): InvokeErrorType separate confusing value/type space 2026-04-28 15:57:51 +08:00
Simon dc41969503 feat(models): support gpt-5.4 and refresh recommendations 2026-04-27 21:40:19 +08:00
Simon 15f3af406c fix(llms): remove tool_choice for deepseek models 2026-04-27 20:32:53 +08:00
zzy-life a7cc935453 feat(llms): add transformRequestBody hook and improve prompt assembly (#480) 
* feat(llms): add transformRequestBody hook and refine prompt handling

* docs(website): document transformRequestBody usage

* refactor(extension): keep function-valued config handling consistent in useAgent

* feat: simplify `transformRequestBody`

---------

Co-authored-by: Simon <10131203+gaomeng1900@users.noreply.github.com>
 2026-04-27 19:46:46 +08:00
Simon 66a85c0dd3 feat: option to disable named tool choice 2026-03-20 17:40:16 +08:00
Simon fe82b11285 feat: AK optional 2026-03-19 19:50:05 +08:00
octo-patch 8ad4d19232 feat: add MiniMax model support with temperature clamping 
Add MiniMax (M2.5 / M2.5-highspeed) as a tested model provider:

- Add model-specific patch in utils.ts: clamp temperature to (0, 1]
  since MiniMax API rejects temperature=0, and remove unsupported
  parallel_tool_calls parameter
- Add MiniMax to the tested models list on the Models documentation page
- Add MiniMax configuration example alongside existing providers
 2026-03-15 23:42:30 +00:00
tsubasakong eec601e6b2 fix(llms): avoid reasoning_effort for GPT-5.4 chat tools 2026-03-08 16:37:32 -07:00
Simon 53db2069ce feat: zod support v3/4 2026-03-05 19:13:18 +08:00
Simon 0dacbda9da chore: improve logging 2026-03-05 16:54:41 +08:00
Simon 68d68182e2 feat(website): drop hash-based router 2026-02-27 19:46:44 +08:00
Simon b4377ed152 chore: remove unused env.d.ts file 2026-02-25 17:19:09 +08:00
Simon b7be29d792 feat(core): enhance tollcall validation with better error reporting 2026-02-25 17:17:35 +08:00
Simon 22fe448d95 refactor: unify zod imports 2026-02-25 16:18:56 +08:00
Simon 49bdc70c78 chore: improve error logging 2026-02-25 16:13:39 +08:00
Simon 8cb57befb3 feat(model): disable thinking for Qwen model patch 2026-02-23 17:30:35 +08:00
Simon dffcb53db9 feat!: Refine lifecycle hooks; fix abortSignal 
- add `stop` method. agent can be reused after stopped
- agent can not be reused after disposed
- extension DO NOT exposes `dispose` anymore. only `stop`.
- update panel for new `stop` method
- fix MultiPageAgent dispose event
- better handling abortSignal
 2026-02-13 17:57:12 +08:00
Simon 1fcf931685 feat(core): add a observe phase in a step; improve abortSignal 2026-02-10 22:20:42 +08:00
Simon 6f0ff1fd33 refactor: zod tree-shaking; better error handling in agent steps 2026-02-09 17:49:10 +08:00
Simon 8ea8ce9acd fix(llms): gpt-5-mini temperature 2026-02-06 17:19:06 +08:00
Simon f6394a04b1 feat(core): add rawRequest in history for debug 2026-01-28 14:30:04 +08:00
Simon dad837e173 fix: don't retry or show AbortError; remove reason from dispose; don't clean history in dispose 2026-01-22 20:19:51 +08:00
Simon 22516dec74 feat: unify InvokeError; add rawResponse to historical ErrorEvent 2026-01-20 20:16:34 +08:00
Simon 1ede1d9911 feat!: redo Panel; decouple Panel from Agent 
BREAKING CHANGES: Agent and Panel API Changes
 2026-01-17 23:14:26 +08:00
Simon 4f6249a252 feat!: require user to provide LLM api. do not fallback to demo LLM 
BREAKING CHANGE: LLM API will be required for agent constructor
 2026-01-14 18:49:23 +08:00
Simon 2f2bdb4697 fix: update Qwen patch condition for model name 2026-01-13 15:07:01 +08:00
Simon 3fa7c8b85a feat: improve qwen3 compatibility (#106) 
* feat: improve qwen3 compatibility

* fix: qwen

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
 2026-01-13 15:04:03 +08:00
Simon 35d1fd1166 chore: better erroring 
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
 2026-01-13 14:16:51 +08:00
Simon 526cb4dbb3 chore: rename OpenAIClient; clean up old notes 2026-01-13 13:58:21 +08:00
Simon 6dc56c57c6 chore: rm legacy llm client 2026-01-13 13:56:30 +08:00
Simon a3263fdd3a chore: rename OpenAICompatibleClient 2026-01-13 13:51:09 +08:00
Simon 14974c0257 feat!: mv brain from llms to agent; redo toolCall auto fixer 2026-01-13 13:49:19 +08:00
Simon bfa6a5347e fix: rm maxTokens from api 2026-01-10 23:27:08 +08:00