RAG / Knowledge Agent

Your company's memory,
available to every agent.

Internal documents, past records, historical logs — the institutional knowledge every team relies on. RAG / Knowledge Agent manages embedding and semantic retrieval over all of it, giving every other agent in your orchestration access to it as shared memory.

Semantic Retrieval Shared Across Agents
Semantic Retrieval

Search by meaning,
not just keywords

Documents, past tickets, and logs are embedded and indexed so a query finds what's actually relevant — not just what shares a literal word. Every other agent can query this collection instead of maintaining its own retrieval logic.

  • Indexes internal documents, historical records, and logs into one searchable collection
  • Retrieval is by semantic similarity, not exact keyword matching
  • Any other agent can query the collection — no duplicated retrieval logic per agent
  • New documents are embedded and made searchable automatically, no manual re-indexing
See It in TotalApp
totalapp.app / em
Knowledge Query
1 Query: "What was our return policy exception for damaged goods?"
2 3 relevant chunks retrieved from Policy Handbook v4
Retrieved context handed to the requesting agent, with citations.
Live in TotalApp
Shared Across Agents

One knowledge base,
every agent draws from it

Rather than each agent maintaining its own retrieval pipeline, they all query this one shared knowledge layer — keeping answers consistent and avoiding the drift that comes from duplicated, independently-maintained indexes.

  • Any agent — Core or Industry-Specific — can query the same knowledge collection
  • Access scoped by tenant and role, same as every other data source in TotalApp
  • Citations included with every retrieval so answers can be traced to their source
  • Works alongside the 2-Tier Architecture for workflows that need both exact and semantic lookup
See 2-Tier Architecture
totalapp.app / sh
Shared Retrieval Layer
Compliance Agent queried the same collection for a policy check
Support Bot queried it for a customer-facing FAQ answer
One index, consistent answers, no duplicated retrieval logic.
Live in TotalApp
Why It Matters

Memory, shared across your whole platform

Every agent draws on the same institutional knowledge — no drift, no duplicated pipelines.

Company-Wide Indexing

Documents, records, and logs embedded into one searchable collection.

One Shared Layer

Every other agent queries the same knowledge base — no duplicated retrieval logic.

Citations Included

Every retrieval comes with a source reference, so answers can be traced and verified.

Tenant & Role Scoped

Access to the knowledge base respects the same RBAC boundaries as everything else.