赵月辉 / fastgpt-migrated

* fix: chat home login

* doc

* feat(app): implement workflow node inline edit and spacing updates

* refactor(app): extract InlineEdit component and implement workflow node inline editing

* feat: persist share chat language preference

* fix: simplify language switching persistence

* doc

* fix: isolate share language preference

---------

Co-authored-by: archer <545436317@qq.com>

* refactor(web): refactor MyModal and migrate its usages

* refactor(app): migrate modals to v2 MyModal with skill dashboard

* refactor(modal): modal spacing; standardize scrollbars

* refactor(modal): clean up custom padding and icon props on MyModal migrations

* upgrade ts6

* fix docs css type declaration

* refactor: extract ChatBox hooks and components

Split ChatBox/index.tsx into hooks, utilities, and UI components while
keeping the public API unchanged. Includes main #6977 stop chat overrides,
mark-read hooks, and quote list type updates.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: make requestVariables time test timezone-agnostic

Use formatTime2YMDHMS for expected values so CI in UTC matches local runs.

Co-authored-by: Cursor <cursoragent@cursor.com>

* refactor: organize ChatBox utility modules

* fix: keep active chat refs in sync before resume callbacks

* test: cover ChatBox utility modules

* perf: ui

* test: stabilize workflow benchmark scaling check

* fix: reason render

* fix: ts

* submodule

* doc

---------

Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: archer <545436317@qq.com>

* test: parallelize mongo-backed vitest suites

* ci: split FastGPT test jobs

* ci: publish one coverage report

* test: fix app suite under retained mongo indexes

* test: keep app vitest config self-contained

* test: preserve root username in fixtures

* feat(app): introduce resource reference index and optimize skill appCount query performance

* refactor(app): standardize api input validation and optimize backfill performance

* fix: tighten skill resource refs backfill

* fix: use version resource refs for skill app count

* fix: cache skill resource refs on apps

* chore: remove unused app version refs index

* perf: optimize skill list permission lookup

---------

Co-authored-by: archer <545436317@qq.com>

* feat: support multimodal LLM inputs

- add model capability config for vision, audio, and video inputs
- wire multimodal flags through simple app, chat agent, and workflow LLM nodes
- normalize uploaded files and file links into image/audio/video request parts
- adapt chat, agent, and toolcall runtime context for multimodal messages
- add capability tags, i18n entries, design docs, and related tests

* perf: code

* test: update agent file prompt assertion

---------

Co-authored-by: archer <545436317@qq.com>

* doc

* docs: translate cloud FAQ

* doc

* perf: count token

* doc

* perf: quote code

* fix: type

* doc

* doc

* add zod schema

* perf: deploy

* doc

* noproxy

* perf: deploy

* doc

* deploy

* perf: deploy

* doc

* default editor size

* perf: deploy

* doc

* refactor(sandbox-proxy): add subdomain proxy auth foundation

Squashes:
- 545b5b767 refactor(sandbox): subdomain-routed proxy with cookie auth and direct execd
- 3bfeb1384 fix(sandbox): align proxy token TTL handling
- 62a642ce8 refactor(sandbox-proxy): consolidate auth/token modules and simplify tests

* style(skill): refine header and background UI

* feat(skill): add version switch

* feat(sandbox-proxy): manage code-server sessions

Squashes:
- e5c07bf5f refactor: refine sandbox-proxy architecture
- d95a7e539 feat: implement code-server session management and update sandbox proxy configuration/permissions
- fe9f1476e refactor: replace manual Map-based cache with lru-cache for session and backoff management

* feat(sandbox): adapt devbox code-server

* fix(sandbox): stabilize debug sessions and sandbox reuse

Squashes:
- c07e2070d fix(debug): route skill preview chat controls through debug session APIs
- 0e251569c fix(sandbox): stabilize agent skill sandbox reuse
- 9f8b672c3 fix(sandbox): split FastGPT sandboxId from providerSandboxId

* refactor(sandbox): clean up provider and proxy routing

Squashes:
- f18361444 feat: sandbox-proxy images and provider cleanup
- dfc27ba3b refactor(sandbox): move resource cleanup into SandboxClient
- 65015efb8 refactor(sandbox-proxy): derive sandbox id from host label

* refactor(sandbox): use stable sandbox instances

Squashes:
- a9f18f873 refactor(sandbox): sandbox to use stable sandboxId
- 3a2ad1691 refactor(db): sandbox DB ops into instance module

* feat(sandbox): preserve folder structure with generated ids

Squashes:
- 8e1374491 fix(sandboxId): use generateSandboxId no matter mode
- 78b5ce27f feat(sandbox): keep sandbox folder struct
- 9f0382305 fix(sandbox): remove unused skill reference params

* fix(skill): tighten sandbox permission handling

* fix(sandbox-proxy): add friendly iframe error fallback

* feat: improve skill sandbox availability checks and prompts

* fix: requestIds required

* feat: sandbox provider handling and proxy revision

* fix(skill): require write permission for edit-debug sandbox proxy token

* fix: type test

* fix(switch): create failed when switch version

* feat(skill): create pending skills asynchronously

* rewrite skill runtime

* fix(switch): create failed when switch version

* feat(skill): skillIds filter and deleted-skill UI

* feat(skill): DeleteConfirmInput and modal support

* feat(sandbox): retry sandbox exec on timeout; UI/i18n updates

* rewrite skill diractory

* feat(skill): use safe skill directory name in edit sandbox & UI label updates

* rewrite skill schema

* refactor(skill): transactional delete, sandbox resource cleanup, and test cleanup

* perf: loading animation

* refactor(app): extract SaveAndPublishModal as common component and support custom title

* fix(skill): align package creation and edit sandbox flow

* perf: api

* fix(skill): tighten skill mutation validation

* feat(skill): introduce inline SandboxEditor & fileOp API, remove legacy sandbox-proxy

* fix(sandbox): support binary file uploading and fix preview corruption in sandbox editor

* test(sandbox): remove obsolete heartbeat test for legacy sandbox proxy

* chore: rm sandbox endpoint types; add runtimeTools

* feat(sandbox): support team plan verification, file optimistic updates and   custom avatar for skill imports

* perf: sandbox code

* doc

* fix(sandbox): fix pro test failure and unify sandbox API auth

- Update pro submodule commit to eacfd364b to solve obsolete agentSkills imports

- Refactor sandbox API endpoints to use Unified authSandboxSession auth

- Support teamId parameter in getSandboxClient and resolve runtime permission check

* perf: stop chat in skill edit

* refactor(sandbox): adjust sandbox permission check and migrate skill APIs to parseApiInput

* refactor(sandbox): extract sandbox files state and actions to useSandboxFileStore hook

* perf: prompt and list recursive

* submodule

* feat: fileOp queue

* doc

* check empty response

* fix(sandbox): prevent duplicate file loading in editor and add hover title for tabs

* show skill

* feat: update sandboxEditor modal

* perf: storage code

* skill ui

* fix(skill): preserve skills root directory structure in sandbox workspace

* fix: adapt

* fix(skill): normalize sandbox zip extraction paths and cleanup unused sandbox checks

* loading ui

* fix: export import and perf loading

* submodule

* fix: repair skill edit sandbox ci

* fix(skill): harden sandbox edit and cleanup flows

* ui

* perf: sealos default workspace

* perf: code

* perf: code

* select skill

* perf: docker secret

* perf: skill editor ui

* fix: test

---------

Co-authored-by: archer <545436317@qq.com>

* fix: sync chat stop state

* docs: add stream resume stop state design

* fix: reset stale generating chats by stream activity

* docs: add stream resume stale reset design

* fix: sync temporary chat history title

* docs: add stream resume history title design

* fix: validate chat data ids before run

* docs: add stream resume data id validation design

* fix: preserve resumed flow node responses

* fix: merge completed resume records with replayed node responses

* fix: keep loaded AI output when resume starts

* fix: skip stale resume interactive after submitted form

* fix: refresh interactive form values during resume

* fix: render form input result files in response detail

* fix: render resumed form input files

* fix: hydrate resumed form interactives

* fix: preserve resumed form file values

* fix: recover resumed form values

* fix chat resume interactive state

* remove stop chat warning toast

* fix: prevent chat history leaking when switching apps

Sync appId/chatId on app switch and align history list state so sidebar no longer shows the previous app's conversation.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: restore last chat per app when switching apps

Remember chatId per app and fall back to the most recent history instead of always starting a new conversation.

Co-authored-by: Cursor <cursoragent@cursor.com>

* docs: add detailed comments for chat resume and form input recovery

Document stream resume merge logic, formInputResult normalization, and chat store storage layering to make the recent form file recovery fixes easier to maintain.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: resolve WholeResponseModal merge artifact and build type errors

Remove the stale monolithic WholeResponseModal.tsx left from a bad merge, wire FormInputResult into the modular response modal, and drop invalid stepId folding logic in ChatItem.

* docs: 41502

* fix: copy server-proxy.js into app image for SHOW_SKILL entrypoint

Restore the runner-stage COPY removed in #6808 so pods with SHOW_SKILL=true can start server-proxy.js instead of failing with MODULE_NOT_FOUND.

Co-authored-by: Cursor <cursoragent@cursor.com>

---------

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: chatitem animation

* fix: chatitem animation

* support image embedding dataset indexes

* feat(dataset): support image embedding system indexes

* fix: migrate legacy dataset search input

* perf: code

* perf: code

* submodule

* add test

* fix: migrate dataset search input

* doc

* test: update dataset search input expectation

---------

Co-authored-by: archer <545436317@qq.com>

* chore: suppress external api zod input noise

* chore: migrate api input parsing

* chore: update pro api input parsing

* fix: stabilize image caption child response id

* refactor: simplify zod error response handling

* fix: set charset for source file previews

* fix: avoid charset on explicit binary source previews

---------

Co-authored-by: xqvvu <whoeverimf5@gmail.com>

When using ali-oss as STORAGE_VENDOR, the `put()` method treats string
  body as a file path and calls `fs.statSync()` on it. Passing `'ok'`
  causes `ENOENT: no such file or directory, stat 'ok'` during
  instrumentation-check, crashing the server on startup.

  Using `Buffer.from('ok')` ensures all adapters (aws-s3, minio, cos, oss)
  handle the body as content rather than a file path.

* fix: fix left-aligned list marker styling in chat welcome guide

* fix: make external S3 endpoint check non-blocking during health check

When using OSS storage with Next.js, getBucketInfo() triggers
"ReferenceError: name is not defined" at startup. This happens because
the ali-oss SDK references a variable named `name` internally in its
createRequest function, which conflicts with the global `name` property
in bundled JavaScript environments.

Replacing getBucketInfo() with list({ 'max-keys': 1 }) achieves the same
goal (verifying the bucket is accessible) while avoiding the problematic
code path. The list() method is already used successfully elsewhere in
the OSS adapter.

Fixes #6552

Co-authored-by: octo-patch <octo-patch@github.com>