Getting started

Documentation

AgentSwarms is a platform for building agentic AI on your own data — agents and multi-agent swarms, the knowledge and tables they work from, the dashboards beside them, and the controls that make all of it safe to hand to other people.

New here?

Read Quickstart with the app open in another tab — thirty minutes, and you'll have an agent answering from your own data. Then Core concepts for the vocabulary.

Start here

Find it by what you're doing

Build something

PageCovers
Agent BuilderPrompt, model, tools, knowledge, memory, guardrails, versions and export.
Agent ChatRun agents, visual BI answers, PowerPoint/Word/Excel generation, sources.
Swarm CanvasMulti-agent graphs: routers, loops, parallel work, approval gates, framework export.
Skills & Prompt LibraryReusable capabilities and prompts.
Developer workspacePython notebooks with real kernels, model access and KB retrieval.

Work with data

PageCovers
Data Catalog & SQLUploads, warehouse connectors, catalog, profiling, lineage, the SQL workbench.
Data preparationVisual joins and transformations saved as repeatable, schedulable flows.
Knowledge BaseIngestion, chunking, embeddings, graph search, and debugging bad answers.
Semantic LayerGoverned metrics so one question has one answer.
BI WorkspaceDashboards, filters, drill-through, alerts, sharing and embedding.

Ship and integrate

PageCovers
IntegrationsConnect providers, search, automation and data sources.
Models & providersBYOK, the model registry, and choosing the right model per job.
MCP serversGive agents tools owned by other systems.
Web embeddingPut an agent or dashboard on your own site, safely.
API & webhooksRun swarms from your code: scopes, idempotency, signed callbacks.
SecretsStore credentials once and reference them everywhere.

Govern and operate

PageCovers
Access controlUsers, groups, model rules, read-only sharing, invite-only signup and SSO.
Guardrails & PIIInput/output controls and personal-data detection or redaction.
Budgets & costSpend caps per user, group, embed or API key — plus retention windows.
Logs & tracesRead what an agent actually did, rather than what it says it did.
Analytics & auditUsage, spend attribution and the audit log.

Three things worth knowing up front

Prose and numbers are different problems. Documents go in the Knowledge Base to be quoted; rows go in the Data Catalog to be counted. Models are poor at arithmetic and excellent at sounding certain, so putting a number where only prose retrieval can reach it is the most common way to get a confident wrong answer.

The trace is the truth. An agent's explanation of its own reasoning is generated text and can be wrong. The trace is a record. Debug in that order.

Your keys, your data. Connect your own model providers and calls run under your account and your agreement with that vendor. See Models & providers.