Proposals nobody reuses
The proposal that won a similar client last year sits in a partner's inbox, and the team writes a new one from scratch.
Memory for Professional Services
Methodologies, delivery templates, approved proposals and checklists become a reviewed knowledge base and skills, with an owner and a review deadline. Each engagement has its own agent, which queries only the documents the firm chooses.
The problem
And every new engagement starts from a blank page.
The proposal that won a similar client last year sits in a partner's inbox, and the team writes a new one from scratch.
People with years at the firm follow the methodology from memory. Someone who just joined builds the deliverable their own way, and the client notices.
Each consultant pastes client documents into a different AI, without knowing what leaves the firm or how much each practice spends.
Before and after
From the approved proposal to delivery in the firm's standard.
| Without the company's memory | With Kluco |
|---|---|
| Without the company's memoryThe approved proposal is in a partner's inbox | With KlucoReference proposals are in the knowledge base, with an owner and the date of their last verification |
| Without the company's memoryEach consultant builds the deliverable their own way | With KlucoThe methodology becomes a skill, and people and agents follow the same steps |
| Without the company's memoryWhat was agreed with the client stays in the notes of whoever was in the meeting | With KlucoThe recorded meeting becomes minutes in the knowledge base, after review, without participants or personal data |
| Without the company's memoryAny AI reads any client's documents | With KlucoEach engagement agent queries only the documents the firm chooses |
| Without the company's memoryNobody knows how much each practice spends on AI | With KlucoEach practice's AI spend shows up in Brazilian reais, with the month's forecast |
What goes into memory
Documents come in from Google Drive, OneDrive, SharePoint, file upload, API or webhook, and recorded meetings become minutes that go through review before entering the knowledge base.
Use cases
The most common use cases at consultancies, accounting and audit firms, engineering firms and agencies, with the part of the platform that does each one.
The steps of a process, the criteria of an analysis or the tone of an answer are kept as a skill, with versions. Agents and people use the same skills.
BrainTemplates, clauses and guidance sit in the knowledge base with an owner and a review deadline. AI uses the current version and shows where it came from.
BrainEach recorded meeting becomes a document with the minutes, without participants and personal data, and goes through review before it enters the base.
BrainEach agent queries only the documents you choose, and the access profile sets who can use each agent.
AgentsEach team's AI spend, in Brazilian reais, with the month's forecast and suggestions for cheaper models.
LLM Management and ChatChatGPT, Claude, Gemini and other MCP-compatible tools query the same knowledge base, each person with their own access.
BrainBrain · the memory
The Brain turns methodologies, templates and approved proposals into articles with an owner, a review deadline and a version history. The steps for each type of deliverable are kept as a skill.
Agents · AI agents
Create an agent for each practice or engagement and choose the documents it may query. One client's material stays out of another client's agent.
LLM Management and Chat
Each person uses the models their profile allows, in the Chat, which draws on the firm's memory. Each practice can be a team, with its own budget and cost centre code.
Control
The firm sets up each control below in the console.
Related
FAQ
Through agents. Each agent queries only the documents you choose, and the access profile sets who can use each agent. An engagement's agent cannot see the documents left out of it.
If you turn that source on. Each recorded meeting becomes a document with the minutes, without participants, emails, phone numbers, CPF or CNPJ, and goes through review before entering the knowledge base.
Yes. ChatGPT, Claude, Gemini and other MCP-compatible tools query the same knowledge base, with the methodologies and templates, each person with their own access. Skills can also be exported in the open SKILL.md format.
Documents come in from Google Drive, OneDrive, SharePoint, file upload, API or webhook. The platform does not replace project management systems: it organises what the firm knows and delivers it to people and agents.
Next step
In the demo, we show the knowledge base of methodologies and proposals, a delivery skill, an agent per engagement and AI cost per practice.