Develop, release, and self-check Recall extensions. Use when working on Recall ext/extension host behavior, official extensions under extensions/recall-, extension manifests, extension catalog and binary release flow, managed install/upgrade/remove design, or docs that explain Recall's extension model.
Use Recall as a project memory layer for local AI coding session history. Trigger when the user mentions recall, wants a live share link (分享会话/分享对话/分享这个对话/给我链接/share session/session link), wants to refresh or update an existing share link (更新分享链接/刷新分享/重新分享/update share link), resume or open a session, search or export…
Use Recall Reflect to review AI coding workflow history as a conversation-first timeline and discuss observed patterns before changing behavior. Trigger when the user asks to reflect on a project, review AI coding workflow, inspect workflow friction, discuss repeated corrections, review handoffs, or explore possible…
Use the gmc CLI for explicit gmc requests: generate commit messages, create or remove worktrees, prune merged worktrees, discover or sync shared resources, or run other gmc wt subcommands. Triggers mainly when the user invokes /gmc or mentions gmc wt add, gmc wt prune, gmc wt share, gmc wt sync, gmc tag, or asks to…
Build source-faithful and Dify-native frontend prototypes for Dify screens and proposed Dify features. Use when the user asks for Dify prototypes, Dify UI screens, Knowledge/Dataset/RAG prototype pages, login/default pages, navigation tabs, modals, tables, creation flows, or any frontend work that must follow Dify…
Analyze, plan, design, and review Dify Dataset, Knowledge, and RAG product work. Use for Dataset lifecycle, Workflow Knowledge Retrieval nodes, field contracts, user stories, and KnowledgeFS-to-Dify product translation.
Load or list repo-scoped and user-scoped agent brains. Use when the user asks to load an agent by name, list available agents, inspect agent descriptions or workflows, or start a session or task with a personal agent brain.