Use RAGFlow as a retrieval and context layer for agent workflows
Build a supervised RAG context layer with RAGFlow so agents can index documents, retrieve grounded context, and answer with traceable source support.
npx skills add agentskillexchange/skills --skill use-ragflow-as-a-retrieval-and-context-layer-for-agent-workflows
Use this skill when an agent/operator needs to turn a corpus of files, knowledge-base exports, or enterprise documents into a maintained retrieval layer before asking an LLM to answer or act. The workflow is to deploy or connect to RAGFlow, ingest and parse documents, configure retrieval/chunking behavior, test grounded answers against source material, and expose the resulting retrieval endpoint or agent template to downstream assistants. Invoke it when normal product search is not enough because the agent needs repeatable, auditable context assembly across private documents. Keep the scope to retrieval-context setup, ingestion validation, and supervised answer grounding; do not treat RAGFlow as a generic app platform or as an autonomous agent replacement.
What this skill actually does
Inputs and prerequisites: RAGFlow, Docker or supported self-hosted/cloud deployment, source document corpus.
Setup notes: Follow the RAGFlow documentation or repository quick start to run the service, then create a knowledge base, ingest documents, tune retrieval settings, and validate grounded answers before connecting agents.
Source and verification boundary: use https://ragflow.io as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.