Skip to content
MCP server

Unl

Stores project decisions with their reasons and gives them to coding agents first

Install
Pricing
Freemium
Platforms
CLI, macOS, Linux
Runs
Remotely
Version
0.2.3
Listing updated
2 days ago

About Unl

Unl is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Unl targets a common gap with coding agents: rules say what you chose, but not why, so the agent cannot apply the choice to a new case. Unl stores each decision with its reason, author, scope and history. Running npx unlimitless init reads your existing CLAUDE.md, AGENTS.md, Cursor rules and ADRs, shows what it found and lets you choose what to keep. When you replace a decision, the old one stays in the history and is marked replaced.

Supported hooks bring relevant decisions in before the agent acts. Claude Code has prompt and pre-edit hooks, Codex has a prompt judge with a full-session proof still pending, and Cursor uses word and file matching. Tools that read AGENTS.md get a generated decisions section. A receipt command shows what was handed over, though the README stresses that delivery does not prove the agent followed it.

The record lives on your machine, in git storage or .unl/local, with readable files in .unl/decisions/. With telemetry off the local package makes no network connection; the optional judge asks your own coding agent and its provider. A separate hosted connection, which needs an account, brings decisions into other AI chats, but it does not yet sync with the local record.

Install

Run in a terminal

claude mcp add --transport http unl https://api.unlimitless.ai/mcp

Add to Settings → Connectors → Add custom connector

https://api.unlimitless.ai/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "unl": {
      "url": "https://api.unlimitless.ai/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "unl": {
      "type": "http",
      "url": "https://api.unlimitless.ai/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "unl": {
      "serverUrl": "https://api.unlimitless.ai/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "unl": {
      "type": "http",
      "url": "https://api.unlimitless.ai/mcp"
    }
  }
}

What it does

Decisions with reasons
Stores wording, reason, author, scope and history together.
Import from rules files
Reads CLAUDE.md, AGENTS.md, Cursor rules and ADRs.
Hook delivery
Gives relevant decisions to the agent before supported actions.
Replace with history
A confirmed new decision supersedes an old one and keeps the trail.
Delivery receipt
Shows what was handed to the agent, without claiming it was followed.
Local record
Keeps the database in git storage or .unl/local.

Good for

  • Stop re-explaining project conventions to each new agent session
  • Capture why a library or tool was chosen
  • Move decisions out of scattered rules files
  • Record a changed decision without losing history
  • Share decisions with other AI chats

Pricing

Freemium

Local package free (Apache-2.0); hosted connection has Free, Basic, Pro, Max and team tiers

Prices change. Check unlimitless.ai before you buy.

Unl FAQ

What does it cost?

The local open-source package is free under Apache-2.0 with no account. A hosted connection has paid tiers, shown on its pricing page without amounts here.

What does it require?

Node.js 20 or later and a coding agent.

Does it send data anywhere?

With telemetry off the local package makes no network connection, though the optional judge sends material to your coding agent's model provider.

Does it guarantee the agent obeys?

No. The README says it delivers decisions but the agent still reasons and you decide.

Is Unl free?

Unl has a free plan, with paid plans for more usage or features.

How do I add Unl to Claude Code?

Run this command in your terminal: claude mcp add --transport http unl https://api.unlimitless.ai/mcp

What are the best alternatives to Unl?

The closest alternatives to Unl are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Unl

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools