FinAIOSby FinoGeeks

Product · Knowledge layer

FinLore

The knowledge base your agents cite

Policies, research notes, client files and the internal wiki become retrievable, citable and privately hosted. Agents do the work; FinLore owns what they read and what they cite.

Talk to FDESee in the stack
FinLore hero: answers with their sources

Where it sits

The knowledge layer of FinAI OS

FinLore is the read path for agents. FinClaw does the work, FinGuard governs writes to enterprise systems, and FinSAFE contains how it runs. FinLore decides which corpus an agent reads and what it cites. It is deliberately not a second agent console. Sandboxing and exactly-once writes to enterprise systems belong to FinSAFE and FinGuard, and skill review to FinHub. Fintology is designed to define the rules, schema and ontology that FinLore's knowledge follows.

Sold on its own or integrated with the other FinAI OS products.

FinAI OSThe managed agent platform · on-prem or your private cloud
Layer 4
Experience
Where clients, advisors and staff meet the agents.
FinDeskPlugin-based white-label AI desktopFinClipMini-program runtime for your apps
Layer 3
Agent
Manage, run and contain agents per tenant.
FinHubManaged agent platformFinClawMulti-tenant agent runtime
Layer 2
Knowledge
What agents know, and how it is structured.
FinLoreKnowledge baseFintologyOntology + knowledge graph / distillery
FinLore is here
Layer 1
Control
Every agent runs contained; FinGuard decides each write that goes through it.
FinGuardAction gatewayFinSAFESandbox
Your infrastructure  Models · data · identity · core systemsHighlighted: Knowledge layer

Capabilities

What FinLore does

Five things a firm needs before agents can answer from its own documents.

  1. 1

    Bring the documents in

    Upload files or sync them from the workplace wikis and document tools a firm already uses. FinLore parses PDF, Word, Excel, images and other common formats, and keeps the folder structure of what you upload.

    Multi-source sync10+ formats
  2. 2

    Retrieval you can tune

    Search by meaning, by exact keyword, or both. Teams can inspect how documents are split, edit individual passages, and roll an edit back. Edits are reindexed automatically.

    Hybrid searchEditable passages
  3. 3

    Answers with citations

    Ask a question of a knowledge base and get an answer that points to the passages it came from.

    CitationsQ&A
  4. 4

    A wiki that maintains itself

    FinLore can distil a knowledge base into interlinked wiki pages, with revision history, line-level diffs, in-browser editing and one-click rollback. FAQ entries can be curated alongside.

    WikiRevision history
  5. 5

    A knowledge tool for agents

    Any MCP-capable agent, including FinClaw, can attach to a FinLore workspace. Each endpoint has its own tool allowlist and knowledge-base scope, access is re-checked on every call, and write tools appear only where they have been enabled. Each knowledge base keeps an activity log.

    MCPHTTP APIScoped access

Standalone · better with

Useful alone. Stronger in the stack.

FinLore runs on its own as a private knowledge base with cited Q&A. In the stack, agents connect to it over MCP for what they read and cite.

FinLore alone

FinLore on its own: a private, citable knowledge base

  • Cloud-native containers, on-prem or in your private cloud
  • Documents and vectors stay inside your enterprise
  • Installs from an offline package, with no image registry needed on the target host
  • Bring your own models for chat, embedding and reranking
  • MCP and HTTP API for any agent or application

Better with

+ FinClawMulti-tenant agent runtime
Agents that read the right corpus

Attach a FinLore workspace to FinClaw over MCP. FinClaw does the work, and FinLore controls what it reads and cites. Any MCP-capable agent can attach the same way.

+ FintologyOntology + knowledge graph
Designed to follow your ontology

Fintology (in design) is designed to define the rules, schema and ontology FinLore's knowledge follows, and to publish reviewed content into FinLore so answers come from it.

In the wealth solution

What the advisor agent can cite

In Personal Wealth Agents, FinLore holds IPS documents, product documentation and house views, all citable.

IPS and goals

The agent's answers draw on the client's IPS and household goals.

House views

Investment-office views and research notes that advisors and agents quote from the same source.

Product documents and policies

Product documentation and policies the agent can cite when it explains a proposal.

See Personal Wealth Agents →

Client app and licensed advisor console shown on the solution page.

Deploy with FDE

Your knowledge, made citable

Forward Deployed Engineers work with your content owners to bring documents in, tune retrieval, and connect your agents to the right knowledge bases, on-prem or in your private cloud.

Talk to FDE
A typical FinLore engagementTypical sequence · scope varies
  1. Step 1

    Assess

    Deploy FinLore on your infrastructure, offline if needed, and connect your chat, embedding and rerank models.

  2. Step 2

    Ground

    Sync sources, set parsing and chunking per knowledge base, and check retrieval on real questions.

  3. Step 3

    Hand over

    Publish scoped MCP endpoints so each agent sees only the knowledge bases and tools it should.

Next step

Ground every answer in what your institution actually knows

Talk with a Forward Deployed Engineer about running FinLore on its own or as part of FinAI OS.

Talk to FDEExplore FinAI OS