Build on RAGfly
Give your agent the exact
document context it needs.
RAGfly exposes your group's document corpus to any external system — agents, code, automations. Five interfaces, one auth contract, multi-tenant RBAC by design. Pick a path below or grab the raw docs.
Before you start — get your API Key
The group admin creates a key at app.ragfly.ai/api-keys (or via POST /auth/api-key). Shown only once — store it in a secrets manager.
Authorization: Bearer slm_live_xxxxxxxxxxxxxxxxxxxxxxxxThe five interfaces
Pick the one that fits your stack — all share the same auth contract and the same RBAC.
📦
Python SDK
pip install ragfly. Fastest path from Python: client.ask("…").
🤖Start here
MCP
LLM agents (Claude Code, Cursor, Cline, Codex). The agent discovers the tools itself.
⚡
CLI
Scripts, automations, CI/CD pipelines and terminal diagnostics.
🔌
REST + SSE
Any language or platform: n8n, Make, Zapier, custom apps.
🌐
Web
End users operate at app.ragfly.ai directly — no integration needed.
Guides & reference
How to start, how each runtime behaves, and how to report your evaluation.
🚀
Quickstart
From zero to first semantic query: sign up → API Key → MCP → result.
🧭
Integration guide
Credentials, roles, identity types and the contract common to every interface.
🎛️
Runtime hints
Which tool to use per runtime: short-context agents, reasoners, IDEs, REST.
📄
AGENTS.md
Drop it in the root of your agent workspace (Codex/Claude).
✅
Evaluation template
Report findings from your test with a standard format.
Raw files for agents
Every doc is also served as plain Markdown so an agent can fetch it directly. Point your agent at:
← Back to site© 2026 RAGfly