kōdlo - End-to-end sales workspace with unified context
by•
kōdlo finds prospects that fit your ICP and tells you how to warm each one up. Upload product catalogues and other docs to the Knowledge Hub, and its agent turns them into personalized messages, offers, deep searches and decks. Track client data, collaborate with your team, and manage tasks, all in one place. Connect Claude Desktop/ChatGPT Work or your code agent with kōdlo MCP to scale and orchestrate contexts.
Replies
If I ask it to find new leads, can I tell it to skip companies I already talked to last month?
@hemendra_khatik yes, absolutely. Just tell the agent to skip companies you talked to last month and it will leave them out.
Plus, when kōdlo builds a new prospect list, every one gets a status, so you can see right away who is new and who you’ve already worked with. Nothing slips back into your outreach by accident.
Thanks for the great question!
Congrats on the launch 🐍! The shared memory idea is what stands out to me. Sales prep usually falls apart because context gets lost between tools, so it’s good to see someone fix that at the root!
@anastasiia_ihnatenko Thank you! 🐍 We agree, that’s exactly the problem we set out to solve. kōdlo is the place where all your sales context lives, so you don’t lose it between tools and can scale your sales without starting from scratch every time.
Hey Andriana, huge congratulations on the launch to you and the entire team! 👏
Sales teams often need different offers and messaging for different clients and prospects, and keeping all of that synced and unified is especially valuable when you're working with a sales team. Even as a solo operator, sales people can struggle to keep the right offers and outreach templates in place, particularly for outbound, so kōdlo feels genuinely useful.
The MCP integration is also especially interesting. Being able to connect kōdlo’s context to AI tools like Claude Desktop, ChatGPT Work, and coding agents makes the workflow much more powerful.
Wishing you and the team all the best! :)
@rohanrecommends thank you so much Rohan, and thanks for all your support along the way! 🙌
You touched on something important. kōdlo has its own AI layer in the architecture, so AI isn’t a feature added on top, it’s built into how the whole product works. That also gives every salesperson traceability. You can always see where a piece of information came from and why the agent suggested something.
And that’s why MCP matters so much to us. People already sell with their own AI tools every day. kōdlo collects all the context about your leads in one place, so whether you work in Claude, ChatGPT or a coding agent, you take that context with you and can scale the selling you’re already doing.
It’s also a place where your whole sales team and their agents work together. Through MCP, for ex. Claude can even read the comments your teammates or their agents left in your shared workspace.
@maria_katunina Thanks! The context is built dynamically for each search request by the Discovery agent, using open-source data and data from our providers. The kōdlo agent then uses that context to draft several options for personalized outreach. These usually focus on the prospect’s pain points and how the campaign’s offer can address them.
That said, it’s flexible: you can define a different outreach strategy in the campaign overview, for example opening with industry-specific news. In our experience, other signals like funding rounds, hiring, or job changes are more useful for qualifying prospects than as outreach hooks.
It’s also worth noting that Claude Desktop, when connected, works with kōdlo’s tools a bit differently.
the MCP pivot story in your forum post is the real insight here, not the architecture. "why not just use Claude/ChatGPT" is the question every wrapper-shaped tool gets, and the honest answer is usually "because we hold state those tools don't." one thing I'm curious about: when two people on the same team both connect via kōdlo MCP and work the same account at the same time from separate Claude sessions, does the shared memory merge their changes live, or does one session's context go stale until it reconnects?
@galdayan Happy to answer! When you create a kōdlo account, you also create a domain. As the domain owner, you can invite team members, and each of them can then connect to kōdlo via MCP using their own account.
A domain has both shared and private context. Shared entities, such as the domain overview, campaigns, contacts, tickets, and contact timeline data, are available to the whole team and to any agents they connect. Other settings, like tone of voice, are private to each user. In the Knowledge Hub, sharing is optional, so you decide what’s shared and what stays private.
Any update to the shared context becomes available to all team members right away, including through MCP.
@ivan_telnov that covers propagation, which answers half of it. the part I was actually poking at is conflict, not visibility - if two people on the same domain both edit the same contact's notes within the same minute from two different MCP sessions, does one write just overwrite the other, or is there something that catches the collision and asks a human to pick. propagating fast doesn't matter much if the faster write silently wins.
@galdayan Fair point, and it's the right one to ask. Most of what a team writes in kōdlo doesn't collide by design. Timeline events, tickets and comments are separate entries, each written under the account of the person (or their agent) who created it. Two teammates logging a call on the same contact at the same moment get two entries, not one overwritten note. Editing is also tied to ownership: an event is edited by its author, and a ticket by its owner, with the assignee only able to move its status.
The place where it can happen is the contact's own fields, which any member can edit. Updates there are per field, so one person changing the title and another changing the bio both stick. If two people change the same field at the same moment, the later write wins in the database. We've treated that as a low risk because contact fields don't change often compared with the timeline. But you're right that agents make the window between reading and writing longer than a person filling in a form, so optimistic locking on contact fields is a good idea. We'll look at it.