SOLVED
Your AI needs a backend
AI agents write. AI agents read. But where does the result go, and who enforces the rules? Smeldr is the structured, governed layer where agent output becomes content your team and your audience can rely on.
Get started
No dashboard required.
Your AI handles publishing through a structured API. Your team stays in the tools they already use.
Full control, zero risk.
Role-based access means your AI can only do what you allow. Draft content never leaks.
Live in minutes.
Connect once. From that point, a conversation is all it takes to publish.
An AI agent can write a post. It can draft an update, generate a summary, produce a structured record. The question is what happens to that output.
In most setups, it lands in a document, a message, or a file. Someone has to pick it up, review it, and decide what to do with it. The agent has no way to know whether its output was used, revised, discarded, or published unchanged. It cannot check. It was not built to check.
Smeldr is the layer that closes that loop.
Structure the agent can operate on
When an agent creates content in Smeldr, it goes through the same lifecycle as anything a human creates. Draft first. Published when approved. Archived when done. The agent cannot skip steps. It cannot publish content that has not gone through the lifecycle. The rules are the same for everyone.
This is not a constraint on what the agent can do. It is what makes agent output reliable. Your team knows that anything in Published state has been through the process: regardless of whether a human or an agent created it.
One backend, four audiences
The same content model serves your website, your API, your AI agent, and any AI reading your published content. One set of rules. One place to manage it. An agent creating content and an agent consuming content use the same backend, governed the same way.
Your content is already structured for AI to find, read, and cite. By default. Adding an agent that writes to that same structure does not require a new system: it is the same system, with a new participant who follows the same rules.
The agent knows its limits
Access in Smeldr is role-based. An agent operating at author level can create and publish content. It cannot delete records. It cannot access content it has not been given access to. The access model is the same for agents as for humans: not a separate AI permission layer bolted on top, but the same role system applied consistently.
You can give an agent precisely the access it needs. No more.
*See Your agents work within rules for how the role system applies to AI agents.* *See forge-agent for wiring an agent runtime into your Smeldr app.*
Ready to put your AI to work?
Forge is open source and self-hosted. Get started in minutes.