doc: add comprehensive skills coverage#223
Draft
esafwan wants to merge 16 commits into
Draft
Conversation
Add _index.json with 11 skills covering: - Core features: agent-system, knowledge-system, tool-system, trigger-system, flow-engine - UI components: frontend-core, frontend-chat, frontend-flow-builder - Infrastructure: provider-system, security, observability Each skill includes id, name, category, key_files, depends_on, and tags
Document Core Agent System including: - Agent DocType structure and configuration - AgentManager and run_agent_sync execution flow - ConversationManager with persistence strategies - Agent Chat and Console interfaces - Permission system (guest, user, role binding) - Extension points for custom tools, providers, hooks - 15 gotchas covering async safety, DB commits, token limits
Document Knowledge System (RAG) including: - Knowledge Source and Knowledge Input DocTypes - SQLite FTS5 backend with BM25 ranking - Ingestion pipeline with extractors (PDF, DOCX, HTML, URL) - SentenceSplitter chunking strategy - Mandatory vs Optional knowledge modes - Agent integration via context_builder - Extension points for custom extractors and backends - 8 gotchas covering query escaping, Redis locks, file paths
Document Tool System & MCP Client including: - Agent Tool Function DocType with 19 tool types - SDK Tools bridge to agents SDK FunctionTool - Tool registry with huf_hooks discovery - MCP client for external tool providers - SSRF protection in http_handler - Standard tools: ocr_document, generate_image, generate_audio - Extension points for custom functions and MCP servers - 10 gotchas covering naming constraints, TTS resolution, tool limits
Document Provider & LiteLLM Integration including: - AI Provider and AI Model DocTypes - RunProvider central routing layer - LiteLLM provider with 100+ LLM support - Model name normalization (gpt-4-turbo → openai/gpt-4-turbo) - API key management with environment variables - Multi-turn tool calling and streaming support - OpenRouter provider with retry logic - 10 gotchas covering temperature restrictions, rate limits, key storage
Document Trigger System including: - Agent Trigger DocType with 5 trigger types - Doc Event triggers with safe_eval conditions - Agent hooks with Redis-based duplicate prevention - Scheduler with Hourly/Daily/Weekly/Monthly/Yearly intervals - Webhook triggers with authentication - SSE streaming at /huf/stream/<agent_name> - Frontend: TriggersTab, TriggerModal, TriggerFieldsRenderer - Extension points for custom trigger types - Gotchas covering safe_eval limits, cache expiry, job deduplication
Document Flow Engine (v0.1) including: - Flow Definition DocType with JSON graph storage - Flow Run DocType for execution state - Core modules: flow_engine, flow_api, flow_eval, flow_orchestrator - 6 node types: trigger.webhook, agent.run, tool.call, router.llm, human.approval, end - 4 edge types: always, on_success, on_failure, expression - AST-based safe expression evaluation - Normal vs Agentic execution modes - Conversation modes: flow_shared, node_isolated, agent_default - 10 gotchas covering hop limits, context persistence, approval flows
Document Security & Permissions including: - SSRF protection with private IP blocking - Password field encryption for API keys - Environment variable handling for sensitive keys - Role-based access control with 20+ capabilities - PermissionAwareToolRegistry for tool-level security - AST-based safe expression evaluator (flow_eval) - MUTATING_TOOL_TYPES for write operation protection - Extension points for custom validators and RBAC - 8 gotchas covering URL validation, tool permissions, expression limits
Document Observability & Monitoring including: - Agent Run DocType with token and cost tracking - Agent Conversation and Agent Message DocTypes - Agent Run Feedback for quality ratings - Agent Tool Call logging - Dashboard components with GridView and filters - Execution pages: list view and detail view - Flow integration for node-level tracking - Extension points for custom metrics and webhooks - 8 gotchas covering truncation, permission conditions, streaming differences
Document Frontend Core & Architecture including: - React 18 app bootstrap with Vite - Context providers: User, Permissions, Flow, Integrations, Modal - 50+ shadcn/ui components - Services layer for Frappe API communication - Custom hooks for mobile, debounce, infinite scroll - TypeScript strict mode configuration - Socket.io real-time integration - Extension points for pages, services, contexts, UI components - 8 gotchas covering TypeScript strictness, boolean handling, paths
Document Frontend Chat System including: - ChatPageV2 and ChatWindowV2 components - Message components: ChatMessage, ChatMessageList, MessageActions - AI Elements: artifact, audio-player, chain-of-thought, code-block - Real-time streaming with SSE and Socket.io - Artifact system for JSX, web preview, code - Model switching mid-conversation - Conversation management with history - chatApi and streamChatApi services - 10 gotchas covering message merging, scroll behavior, streaming fallback
Document Frontend Flow Builder including: - FlowCanvas with React Flow integration - FlowContext for centralized state management - FlowService with Map-based storage and subscriptions - Node components: TriggerNode, ActionNode, EndNode - Modals: NodeSelection, ActionSelection, TriggerConfig - 9 trigger types, 5 action types, utility nodes - Drag-and-drop canvas with visual connections - Real-time updates via subscription pattern - Extension points for custom nodes and triggers - 10 gotchas covering in-memory storage, TypeScript strictness
Add comprehensive Skills Directory section including: - Table of 11 available skills with categories - Usage instructions for developers and AI agents - Skill categories explanation (features, ui, integrations, patterns) - Cross-references to skill files for detailed context
Add Skills section documenting: - Machine-readable skill directory at skills/ - Table of 11 skills with categories and descriptions - Reference to skills/_index.json for complete catalog - Note that skills are for both developers and AI agents
…ive_skills_coverage
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Commits Created (14 total)
Branch
doc/add_comprehensive_skills_coverageFiles Added / Modified
skills/directory (~4,500 lines)skills/_index.json→ master catalog with dependencies