Unl, your why agent.For every agent you run.
Your AI already knows how to reason. Unl gives it your reason.
Agents can be replaced without restarting the undertaking. Unl keeps your target, what you decided and why, and gives each agent what matters before it acts. Reported outcomes carry the work forward across tools and models.
Measured on real fleets. With Unl, the fleets acted within everything decided on 70 of 90 tasks against 61. Pre‑registered and hashed: read the result. Measured in a coding agent, a typical serve is about 2,000 tokens: only the decisions that bear on the turn. Does Unl use up my context window?
The first pass, measured. When a decision changed partway through the work, agents with Unl needed 3 corrections in 36 tasks. Agents reading a decisions file nobody had updated needed 14, and agents with no file needed 18. Against a file someone kept perfectly, Unl came out even, with nobody keeping it. Frozen before the run and published whichever way it fell. Read the result.
How little it had to say. In a week of real turns in a coding agent, Unl held 343 decisions and handed over a median of 286 tokens a turn. One decision arrived in full and the rest as a single line each, for the agent to open when it needed them. Nothing the agent already had was sent twice.
When the choice of what bears is wrong. Choosing what bears on a turn is a judgement, and judgements can miss. Everything around it is fixed. What you decided stays yours, word for word, and only you can change it. A replaced decision says what replaced it. Every serve is logged, so you can see exactly what your AI was told on any turn. Nothing Unl leaves out is lost, because your AI can always ask for more.
The API is on every tier, on one usage meter. See pricing
The key is in the transcript now. Only rotating it undoes that.
Every example here is a real decision, and the real turn it changed.
Or plug Unl into the agents you already run, over MCP. It works alongside them.
Running an orchestrator? Your orchestrator moves the work. Unl carries your why. Unl for orchestration
Add Unl to the agents you already run
Tell your agent
Read unlimitless.ai/start.md, set up Unl, and show me the decisions you find.Chat surfaces
Coding surfaces
Same decisions, whichever model or tool you use. Also works with Claude Desktop, Cline, Kilo Code, opencode and any MCP client
Unl gives your AI what you decided, and why, when it matters, in every tool you use. See pricingWhat Unl keeps, and how to delete it
A live fleet, running on Unl.
When your agents need a decision, answer in Slack. Unl keeps the why for their next turn.
- Your agent asksKeep the existing library or replace it?
- You answer in SlackKeep it. The team already knows how to maintain it.
- Its next turnThe agent extends the existing library.
Run it in any terminalnpm install -g unlimitlessthenunl
Point it at your own fleet. Run unl in any terminal, open the stream in your Unlimitless portal, or let your agents read it at GET /api/serve-log with your key. See which why reached which agent, and every heads-up Unl raised.
The missing role in the emerging fleet.
Your agents stop re-arguing what you settled.
Unl keeps each decision you make with its reason, and hands your agent only the ones that bear on the task before it acts.
Connect Unl to keep useful thoughts with their why.
Kept with its reasoning, without becoming a rule unless you settle it.
Add Unl to your AIYou stop writing things down for your AI.
| The usual way | Where it stops | With Unl |
|---|---|---|
| Custom instructions in ChatGPT or Claude | Stays inside one app, and applies to every chat whether it's relevant or not | The same decisions reach every AI you use, and only when they bear on the turn |
| CLAUDE.md, .cursorrules, project knowledge | Lives in one repo and one editor, kept up by hand, and goes stale | Your decisions follow you across repos, tools and models, and update the moment you change your mind |
| Memory features | Remember what you said, not why you decided it | Keeps each decision with its reason, and what it replaced |
| Building your own MCP memory server | You host it, configure it and keep it running | Connect once. Unl is the why agent already running |
Unl notices decisions as you make them, and hands each AI the one that matters, when it matters. See the full comparison.
Every agent in your fleet works from the same decisions.
Unl is the infra agent for your fleet, the one that carries your why to every other agent. It never does the work itself.
Your agents never need to message each other. Each one works from the same decisions, so they pull in the same direction.
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
Add Unl to your AISettle one decision, then serve it.
- Add Unl to your AI and settle one real decision in your own words. When it offers to keep it, say yes.
- Mint a key in the portal.
- Serve it. What comes back is your own decision, with its why.
One address for every agent, over MCP, with a REST API beside it.
REST at api.unlimitless.ai with a Bearer key. It only ever grows, so nothing you build on it breaks.
Every type, as JSON Schema.
What bears on a turn, with its why.
Offer a decision; a person says yes.
A file in, its decisions out.
What a serve is composed from.
Keep a composition by name.
One decision, back to where it was settled.
Connect Unl to watch the world against what you decided.
A crossing arrives when the world bears on your position.
Add Unl to your AIYour agent connects in one step and signs in once.
Add https://api.unlimitless.ai/mcp as an MCP server and sign in once over OAuth. No key and no header. Eight tools arrive, and each one declares what it may do, so your agent reads freely and writes only on your word.
A connect, a serve and a write, start to finish.
Everything else, in one place, for you and your agents.
- The full developer reference, every endpoint, field, error and example
- This page in Markdown, for an agent reading the site
- llms.txt, a short index of the site for AI readers
- The live schema, the API's own published JSON Schema, read with your key
Every response, error, scope and worked example is in the full reference.