Skip to content

Staging#324

Merged
ahmad-ajmal merged 104 commits into
mainfrom
staging
Jun 10, 2026
Merged

Staging#324
ahmad-ajmal merged 104 commits into
mainfrom
staging

Conversation

@ahmad-ajmal

Copy link
Copy Markdown
Collaborator

No description provided.

makiroll1125 and others added 30 commits May 19, 2026 14:24
Collapse 3 WebSocket connections into 1 SocketClient and 5 React contexts
into 13 domain slices. Settings tabs cache server data with hasLoaded
flags so remounts and cross-page consumers don't re-fetch.
makiroll1125 and others added 29 commits June 4, 2026 07:21
* new coding skill: find-skills

* new coding skill: firecrawl (scrape the web)

* new coding skill: excalidraw diagrams

* Update contributing md with SOPs

* new coding skill: planetscale postgres

* new coding skill: planetscale mysql

* new coding skill: planetscale vitess

* new coding skill: planetscale neki

* new coding skill: cloudflare workers best practices

* new coding skills: essential Trail of Bits skills + modified existing skill files to reference CraftBot

* new coding skill: awesome-copilot git-commit

* new coding skill: shannon pentester

* clear conversation and task data with clear command

* improvement:more github action

* Added more google calendar actions

* action expansion for gmail, gdrive, and outlook

* action expansion for Notion, Discord, and Slack

* action expansion for Lark

* action expansion of Jira, Line business, and telegram bot

* action expansion for whatsapp

* action expansion google docs

* action expansion lark calendar

* action expansion twitter

* Lint and Formating Fix

* integration connection in onboarding step

* refactor: migrate browser frontend state to Redux Toolkit

Collapse 3 WebSocket connections into 1 SocketClient and 5 React contexts
into 13 domain slices. Settings tabs cache server data with hasLoaded
flags so remounts and cross-page consumers don't re-fetch.

* selectEnabledSkills

* revamp: modal component

* current version fix

* Check update fix

* Remove TUI support

* fix: display user-invoked skills in Skills icon in Dashboard

* Feat/slash command autocomplete (#268)

* feat: add base slash command autocomplete logic

* feat: add core logic and ui for slash command autocomplete

* feat: add slash command autocomplete for skill invocation

* include command as selectable too and tab to autocomplete function

* feat: add support for autocomplete to use Redux for commands and skills

* feat: fixes for slash command autocomplete

* feat: scroll autocomplete dashboard using up/down arrow keys and better autocomplete using tab/enter keys

---------

Co-authored-by: CraftBot <craftbot@craftbot.dev>

* fix: skill information (invocations, top skills) persist across sessions

* fix: skill information (invocations, top skills) persist across sessions + (upload missing file)

* Reset setting config

* fix: add support for OpenAI Images 2.0 and Gemini Nano Banana 2 image generation

* fix: Gemini is default provider if both OpenAI and Gemini API keys are provided

* feature:added bedrock as provider and fixed gemini and openrouter cache logic

* update requirement and setting of Bedrock

* Improvement/task interative update (#287)

* Improvement:upgrade task panel visual

* improvement:refactor UI code

* feature:improve visual and interactive of action

* feature:added action renderer

* feature:initial mascot update

* mascot animation update

* mascost display action

* git status

* set default zoom level and task and actions panel width

* Refactor of the mascot display screen logic

* Reset setting config

* mascot action display refactor

---------

Co-authored-by: CraftBot <craftbot@craftbot.dev>

* removed repeated validation check

* create dm channel struct check

* Fix issue 238: reinitialize cache session mid task provider switch

* bug:fix tui module not found issue

* UI:task list scrolling logic

* UI update: attachement preview

* UI improve:Increase contrast and other improvement

* UI update:link color and marketplace icon

* add base logic for integration dashboard icon

* fix: dashboard integration not showing data properly

* V1.3.2 Lint Checks and fixes

* feat: base working storage for integrations

* fix: github and other integrations not counting

* feat: add integration icon and analytics in dashboard + finish testeing Discord, GitHub, Gmail, Google Calendar, Google Docs, Google Drive, LinkedIn, and Notion

---------

Co-authored-by: ahmad-ajmal <ahmadajmal1514@gmail.com>
Co-authored-by: CraftBot <craftbot@craftbot.dev>
Co-authored-by: イツミネ <tham_yikfoong@outlook.com>
* pdf actions new version

* Fix banner title centering and edit_pdf accent colour description

* fix: derive subtitle colour from FORMAT.md theme in create_pdf
…hat as user create living UI and submitting the form
* improvement: base logic for image interface

* improvement: address image-gen PR review + onboarding/settings fixes

Resolve the 6 review comments (DRY reinitialize, provider fallback, usage reporting, reinit-on-save rollback, single Gemini REST client via GeminiClient.generate_image). Wire onboarding's chosen provider to image gen, show registry-default models in settings instead of blank, refactor onboarding _complete, and bump openai/anthropic floors.

---------

Co-authored-by: ahmad-ajmal <ahmadajmal1514@gmail.com>
'google' isn't a single integration - Workspace apps are separate (gmail, google_drive, …). Stop advertising 'google' in connect_integration, normalize aliases, reject the 'google' umbrella with guidance, and list valid ids on unknown so the agent self-corrects.
In frozen mode PROJECT_ROOT is the per-user data dir, so the default.png lookup missed the copy bundled in sys._MEIPASS - serving 404 and a broken avatar. Fall back to the _MEIPASS-bundled default when the user-data copy is absent.
#316)

* added video generation interface, and openai, gemini, and byteplus as providers

* feature: Add video gen model toggle in Settings --> Models

* fix: video generation for commented changes on PR#316

* fix: Usage reporting, Key exposure, and N x timeout changes for video and image gen interface

* change gemini provider to use x-goog-api-key

* personGeneration fix

---------

Co-authored-by: CraftBot <craftbot@craftbot.dev>
Co-authored-by: Tobias Garcia <iguana3000tg@gmail.com>
Co-authored-by: ahmad-ajmal <ahmadajmal1514@gmail.com>
@ahmad-ajmal ahmad-ajmal merged commit 6c3fac8 into main Jun 10, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants