Lean Agent Protocol

MCP and API docs
optimized for AI

Repetitive MCP tool calls waste 73% of their tokens. API spec prose and duplications waste 70% of theirs.
LAP compresses both leading to 1.5x faster MCP, 2.4x faster API tasks, and a fraction of the cost.

How It Works

Two paths. Same result: fewer tokens.

MCP Path

Make MCP leaner without touching it

Drop a proxy in front of any MCP server. Tool descriptions, results, and updates get compressed at runtime.

Any MCP Server
LAP Proxy
Compressed tools + results 73% fewer tokens
npx ultra-lean-mcp-proxy install
View on GitHub
API Path

Turn API specs into agent fuel

Compile OpenAPI, Postman, or GraphQL into typed .lap contracts. 70% fewer tokens, zero guesswork for your agent.

OpenAPI / Postman / GraphQL
LAP Compiler
.lap file 70% smaller
Your Agent

Full support for Claude and Cursor

Why LAP

Your agent reads 3,000 tokens. It needs 200.

Every API call starts with a wall of prose. Descriptions, examples, markdown formatting -- all designed for human eyes. Your agent pays for every token.

OpenAPI Spec ~3,000 tokens

# Create a Payment Intent
#
# A PaymentIntent guides you through the process
# of collecting a payment from your customer.
# We recommend that you create exactly one
# PaymentIntent for each order or customer session
# in your system...

POST /v1/payment_intents

# Parameters:
# amount (required) - Amount intended to be
#   collected. A positive integer representing
#   how much to charge in the smallest currency
#   unit (e.g., 100 cents to charge $1.00)...
# currency (required) - Three-letter ISO code
# payment_method_types (optional) - The list of
#   payment method types that this PI is allowed
#   to use. If not provided, defaults to...

LAP Format ~200 tokens

@endpoint POST /v1/payment_intents
@desc Create a payment intent
@required {amount: int, currency: iso4217}
@optional {payment_method_types: [str]}
@returns(200) {id: str, amount: int, status: str}
@errors {400, 401, 402}

Proven compression across formats

OpenAPI
5.2x
Median Compression
Postman Collections
6.1x
Median Compression
AsyncAPI
2.2x
Median Compression
Protobuf
1.8x
Median Compression
GraphQL SDL
1.4x
Median Compression

Complete token efficiency suite

ultra-lean-mcp-proxy

Transparent MCP proxy for runtime optimization. Wrap any server, save 73% tokens, 32% faster, zero code changes.

npx ultra-lean-mcp-proxy install

LAP Registry

Discover and publish token efficient API specs. Browse 1000+ APIs with 70% average size reduction.

1000+ APIs