Core Agents

Rather than one monolithic backend, TotalApp is built from a number of specialised agents. Each agent owns one responsibility — data modeling, workflow execution, integrations, AI orchestration, UI rendering, or observability — so it can be reasoned about, scaled, and evolved independently. Agents are split into two groups: Core Agents that run server-side and do the heavy lifting, and Industry-Specific Agents that turn what the core produces into judgment calls for a given vertical.

What Are Core Agents?

Core Agents are the platform's domain-agnostic backend agents — reasoning units that any module, screen, or add-on can call without their pipeline, input contract, or output shape changing based on which industry is calling them. They own the platform's shared infrastructure: reading unstructured content, applying a configurable scoring or review pipeline, and handing back a structured, auditable judgment.

This is a platform-first framing rather than a feature-first one: Core Agents are not about any single industry's terminology or workflow — they are the general-purpose reasoning layer that every vertical add-on is built on top of.

In one sentence

Core handles the platform's data modeling, orchestration, and workflow infrastructure; Industry Agents turn that infrastructure into vertical-specific judgment calls.

Core Agents Available Today

Two Core Agents are live in TotalApp today:

Matrix Agent

Turns unstructured or semi-structured text — a vendor profile, a sales lead's notes, a compliance document, a CV — into a repeatable score between 0 and 100 against a configurable rule set, plus a written summary explaining the result.

Audit Agent

Reviews documents and brand assets for compliance — checking a file or asset against a defined rule set and producing a findings report a human then confirms or overrides.

Both share the same shape: a deterministic, configurable rule/prompt layer, a structured output contract, and a human-in-the-loop review step — which is what makes them Core rather than tied to one vertical's rules.

Core Agents vs. Industry-Specific Agents

TotalApp's agents fall into two groups:

Core Agents

Own the platform's core infrastructure — data modeling, orchestration, and workflow logic. Domain-agnostic and usable by any module or add-on: Matrix Agent and Audit Agent keep the same pipeline, input contract, and output shape regardless of which screen or industry calls them — only the rules and prompts change per screen.

Industry-Specific Agents

Turn what the core produces into a judgment shaped by one vertical's operational reality. Patent Agent is the first of these — its reasoning, terminology, and the tools it calls are specific to IP & R&D workflows rather than reusable across every add-on.

Frequently Asked Questions

What makes an agent a "Core Agent" rather than an "Industry Agent"?

A Core Agent's pipeline, input contract, and output shape stay the same no matter which screen or industry calls it — only the configured rules and prompts change. An Industry-Specific Agent is built for one vertical's operational reality, so its reasoning, tool access, and approval flow are shaped by that industry's own rules and terminology.

Which agents are Core Agents today?

Matrix Agent and Audit Agent are TotalApp's Core Agents. Patent Agent is currently the platform's only Industry-Specific Agent.

Do Core Agents ever act without human approval?

No — the pattern is human-in-the-loop for anything consequential: the agent proposes a judgment, and a person reviews and confirms or overrides it before it becomes final, the same pattern Audit Agent already uses for its findings reports.

Where do I go to read about a specific agent?

See Matrix Agent for rule-based scoring, Audit Agent for document/brand compliance review, and Patent Agent for the IP & R&D reasoning endpoints.