# Orchid Documentation Orchid is a generic, platform-agnostic multi-agent AI framework built on LangGraph (Python) and Next.js. Audience: developers building or deploying multi-agent AI systems. Each entry below is a documentation page with its URL and description. Last generated: 2026-11-20 ## Overview Ecosystem Overview — Five packages, one coherent framework: library, API, CLI, MCP gateway, and frontend. https://orchestratorindex.com/ecosystem (markdown: https://orchestratorindex.com/ecosystem.md) Quickstart — Run Orchid locally in under ten minutes with the basketball demo. https://orchestratorindex.com/quickstart (markdown: https://orchestratorindex.com/quickstart.md) ## Concepts Agents — How OrchidAgent works: YAML configuration, skills, MCP tools, and the run lifecycle. https://orchestratorindex.com/concepts/agents (markdown: https://orchestratorindex.com/concepts/agents.md) Supervisor — The LangGraph supervisor: routing, history limits, and sliding-window summarisation. https://orchestratorindex.com/concepts/supervisor (markdown: https://orchestratorindex.com/concepts/supervisor.md) Hierarchical RAG — OrchidRAGScope, the 5-level hierarchy, retrieval strategies, ingestion pipelines, and the parse-once pattern. https://orchestratorindex.com/concepts/rag (markdown: https://orchestratorindex.com/concepts/rag.md) MCP Integration — Connecting MCP servers to Orchid agents: none, passthrough, and OAuth auth modes. https://orchestratorindex.com/concepts/mcp (markdown: https://orchestratorindex.com/concepts/mcp.md) OAuth & Dynamic Client Registration — How auth.mode: oauth triggers RFC 9728 → 8414 → 7591 discovery and manages per-user tokens. https://orchestratorindex.com/concepts/oauth (markdown: https://orchestratorindex.com/concepts/oauth.md) Embeddings — Supported embedding models, dimension differences, and re-indexing requirements. https://orchestratorindex.com/concepts/embeddings (markdown: https://orchestratorindex.com/concepts/embeddings.md) Multi-LLM Support — Provider agnosticism via LiteLLM: configure any model per agent, swap at runtime. https://orchestratorindex.com/concepts/multi-llm (markdown: https://orchestratorindex.com/concepts/multi-llm.md) Chat Persistence — OrchidChatStorage ABC, the built-in in-memory default, and durable SQLite/PostgreSQL backends via plugins. https://orchestratorindex.com/concepts/persistence (markdown: https://orchestratorindex.com/concepts/persistence.md) Tool Call Strategies — The all, sequential, and llm_decides strategies and how to register custom ones. https://orchestratorindex.com/concepts/tool-strategies (markdown: https://orchestratorindex.com/concepts/tool-strategies.md) Document Parsing — PDF and image parsing with the parse-once pattern and the pluggable parser registry. https://orchestratorindex.com/concepts/document-parsing (markdown: https://orchestratorindex.com/concepts/document-parsing.md) Mini-Agents — Per-agent opt-in fan-out: decompose a multi-part request into parallel sub-agents, each running its own focused agentic loop. https://orchestratorindex.com/concepts/mini-agents (markdown: https://orchestratorindex.com/concepts/mini-agents.md) Pollen + Bloom — Event-driven AI jobs: Pollen ingests and persists signals; Bloom matches them to triggers and runs them through the supervisor. https://orchestratorindex.com/concepts/pollen-bloom (markdown: https://orchestratorindex.com/concepts/pollen-bloom.md) Chat Summarization — Running summaries, structured entity extraction, RAG-augmented conversation memory, and configurable truncation strategies. https://orchestratorindex.com/concepts/chat-summarization (markdown: https://orchestratorindex.com/concepts/chat-summarization.md) Decision Models — A dedicated LLM layer that returns typed, calibrated judgements for internal routing, gating and selection seams — with deterministic fallbacks. https://orchestratorindex.com/concepts/decision-models (markdown: https://orchestratorindex.com/concepts/decision-models.md) Data Pipelines — Typed, run-scoped DAGs that combine deterministic code, LLM jobs, agents, decisions, maps and sub-pipelines over a validated information graph. https://orchestratorindex.com/concepts/pipelines (markdown: https://orchestratorindex.com/concepts/pipelines.md) ## Packages orchid — The core Python library: ABCs, agents, graph builder, RAG, and persistence. https://orchestratorindex.com/packages/orchid (markdown: https://orchestratorindex.com/packages/orchid.md) orchid-api — FastAPI server exposing the Orchid framework over HTTP with SSE streaming. https://orchestratorindex.com/packages/orchid-api (markdown: https://orchestratorindex.com/packages/orchid-api.md) orchid-cli — Command-line interface for interacting with Orchid without running the API. https://orchestratorindex.com/packages/orchid-cli (markdown: https://orchestratorindex.com/packages/orchid-cli.md) orchid-mcp — Node.js MCP gateway that exposes Orchid to any MCP-capable host LLM. https://orchestratorindex.com/packages/orchid-mcp (markdown: https://orchestratorindex.com/packages/orchid-mcp.md) orchid-frontend — Next.js 16 multi-chat UI with SSE streaming and file upload support. https://orchestratorindex.com/packages/orchid-frontend (markdown: https://orchestratorindex.com/packages/orchid-frontend.md) orchid-storage-sqlite — SQLite storage plugin: chat persistence, MCP stores, event storage, migrations, and the sqlite checkpointer. https://orchestratorindex.com/packages/orchid-storage-sqlite (markdown: https://orchestratorindex.com/packages/orchid-storage-sqlite.md) orchid-storage-postgres — PostgreSQL storage plugin: chat and config persistence, MCP stores, events, checkpointer, and visibility fragment. https://orchestratorindex.com/packages/orchid-storage-postgres (markdown: https://orchestratorindex.com/packages/orchid-storage-postgres.md) orchid-rag-qdrant — Qdrant vector and doc-store plugin for the Orchid AI framework. https://orchestratorindex.com/packages/orchid-rag-qdrant (markdown: https://orchestratorindex.com/packages/orchid-rag-qdrant.md) orchid-rag-chroma — ChromaDB vector-backend plugin for on-disk, zero-infrastructure RAG. https://orchestratorindex.com/packages/orchid-rag-chroma (markdown: https://orchestratorindex.com/packages/orchid-rag-chroma.md) orchid-rag-neo4j — Neo4j graph-store plugin for knowledge-graph RAG. https://orchestratorindex.com/packages/orchid-rag-neo4j (markdown: https://orchestratorindex.com/packages/orchid-rag-neo4j.md) ## Configuration Configuration Atlas — Searchable table of every configuration key across orchid.yml and agents.yaml — filter by name, type, or description. https://orchestratorindex.com/configuration (markdown: https://orchestratorindex.com/configuration.md) Infrastructure Configuration — Detailed reference for every orchid.yml property — runtime, auth, LLM, RAG, storage, MCP, tracing. https://orchestratorindex.com/configuration-reference/infrastructure (markdown: https://orchestratorindex.com/configuration-reference/infrastructure.md) Configuration Reference — Detailed reference for every configuration property in orchid.yml and agents.yaml — split by file with searchable indexes and deep-links. https://orchestratorindex.com/configuration-reference (markdown: https://orchestratorindex.com/configuration-reference.md) Agents Configuration — Detailed reference for every agents.yaml property — defaults, supervisor, agents, tools, skills, guardrails, events. https://orchestratorindex.com/configuration-reference/agents (markdown: https://orchestratorindex.com/configuration-reference/agents.md) ## Examples Basketball Demo — Minimal end-to-end Orchid example: GenericAgents with built-in tool classes, PostgreSQL storage, cross-agent skills, and an event-driven trivia agent. https://orchestratorindex.com/examples/basketball (markdown: https://orchestratorindex.com/examples/basketball.md) Helpdesk Demo — Custom agent class, cross-agent skills, sequential routing, and event-driven ticket triage. https://orchestratorindex.com/examples/helpdesk (markdown: https://orchestratorindex.com/examples/helpdesk.md) Restaurant Demo — RAG with dynamic injection, multi-agent sequential routing, custom agent class, and event-driven deep research. https://orchestratorindex.com/examples/restaurant (markdown: https://orchestratorindex.com/examples/restaurant.md) Recipes — Zero-infrastructure RAG cookbook example using ChromaDB and Ollama — no Docker, no external services. https://orchestratorindex.com/examples/recipes (markdown: https://orchestratorindex.com/examples/recipes.md) Education Studio — Four-agent teaching workflow with concept extraction, quiz generation, lesson planning, and multi-format export via PDF, DOCX, PPTX, and Markdown. https://orchestratorindex.com/examples/education (markdown: https://orchestratorindex.com/examples/education.md) Examples — Working examples showing different Orchid capabilities and integration patterns. https://orchestratorindex.com/examples (markdown: https://orchestratorindex.com/examples.md) Learning Digest Fan-out — Pollen + Bloom event pattern: a weekly cron emits one signal per learner, each running as that user. https://orchestratorindex.com/examples/learning (markdown: https://orchestratorindex.com/examples/learning.md) MCP Authentication Modes — All three MCP auth modes in one example: none, passthrough, and OAuth 2.0 with dynamic client registration. https://orchestratorindex.com/examples/mcp-auth (markdown: https://orchestratorindex.com/examples/mcp-auth.md) Orchid Experts — Fleet of ten RAG-powered GenericAgent experts covering every Orchid concept — pure YAML + Markdown, zero custom agent code. https://orchestratorindex.com/examples/orchid-experts (markdown: https://orchestratorindex.com/examples/orchid-experts.md) Postgres Storage — Demonstrates the orchid-storage-postgres plugin: asyncpg chat persistence, PostgreSQL checkpointer, and visibility. https://orchestratorindex.com/examples/postgres-storage (markdown: https://orchestratorindex.com/examples/postgres-storage.md) Weather Fleet — Three-agent weather fleet: remote MCP server for live forecasts, built-in safety tools, RAG-backed outfit advice, PostgreSQL storage, and sliding-window chat summarization. https://orchestratorindex.com/examples/weather (markdown: https://orchestratorindex.com/examples/weather.md) Custom Storage Backend — Plug in a custom OrchidChatStorage subclass via the storage.class field. https://orchestratorindex.com/examples/custom-storage (markdown: https://orchestratorindex.com/examples/custom-storage.md) Data Pipeline — Direct typed DAG example: deterministic, LLM, agent, decision, map, sub-pipeline and emit jobs over a synthetic customer-health graph. https://orchestratorindex.com/examples/data-pipeline (markdown: https://orchestratorindex.com/examples/data-pipeline.md) Event-Driven Data Pipeline — Bloom-triggered typed pipeline: bearer-authenticated webhook, cron schedule, SQLite event storage and chat-bound results. https://orchestratorindex.com/examples/data-pipeline-events (markdown: https://orchestratorindex.com/examples/data-pipeline-events.md) RAG Retrieval Strategies — Side-by-side comparison of simple, multi_query, hyde, and a custom recency-weighted retrieval strategy. https://orchestratorindex.com/examples/rag-strategies (markdown: https://orchestratorindex.com/examples/rag-strategies.md) Tool Call Strategies — all, sequential, llm_decides, and a custom priority strategy in one example. https://orchestratorindex.com/examples/tool-strategies (markdown: https://orchestratorindex.com/examples/tool-strategies.md) Prompt Customization — Every supervisor and agent prompt extension point — assistant_name, routing/synthesis prompts, prompt_sections, and transformer prompts. https://orchestratorindex.com/examples/prompt-customization (markdown: https://orchestratorindex.com/examples/prompt-customization.md) Gallery Curator — A multi-session art gallery curator that remembers artists, exhibitions, and decisions across sessions using Orchid's full conversation summarization pipeline. https://orchestratorindex.com/examples/gallery-curator (markdown: https://orchestratorindex.com/examples/gallery-curator.md) Festival Producer — Multi-agent festival planning with artist booking, logistics, and marketing agents — backed by rag_augmented conversation memory, orchestrator skills, and structured entity extraction. https://orchestratorindex.com/examples/festival-producer (markdown: https://orchestratorindex.com/examples/festival-producer.md) Architecture Review Board — Multi-agent design review with structural engineering, cost estimation, and sustainability consulting — orchestrator skills and rag_augmented memory across design iterations. https://orchestratorindex.com/examples/architecture-review (markdown: https://orchestratorindex.com/examples/architecture-review.md) Graph Knowledge Base — GraphRAG retrieval strategy: graph traversal fused with dense vector search for multi-hop relationship queries. https://orchestratorindex.com/examples/graph-kb (markdown: https://orchestratorindex.com/examples/graph-kb.md) Wiki Document Ingestion — Two-tier corpus ingestion (headered + recursive), hybrid retrieval, and per-tool RAG overrides with the parse-once pattern. https://orchestratorindex.com/examples/wiki (markdown: https://orchestratorindex.com/examples/wiki.md) ## Reference Best Practices — Recommended patterns for production-grade Orchid deployments. https://orchestratorindex.com/best-practices (markdown: https://orchestratorindex.com/best-practices.md) Glossary — Definitions of key terms used throughout the Orchid documentation. https://orchestratorindex.com/glossary (markdown: https://orchestratorindex.com/glossary.md) Contact — Get in touch with the Orchid maintainer or open an issue on GitHub. https://orchestratorindex.com/contact (markdown: https://orchestratorindex.com/contact.md)