All systems operational · v1.0.0

Cut your AI bill
by 71.5×.
Pay us 10% of what you save.

SurfClaw compresses any codebase or document corpus into a traversable knowledge graph — reducing LLM query tokens by 71.5× on average. Zero fixed infrastructure cost. You keep 90% of every dollar saved.

Start saving now →Read the docs

Install the SurfClaw SDK

$ pip install surfclaw
71.5×
average token reduction
$0
fixed monthly server cost
25
languages supported
99.9%
gross margin
▶ surfclaw_demo.py
from surfclaw import SurfClawClient

client = SurfClawClient(api_key="sk-sc-...")
result = client.process_directory("./my-codebase")

print(f"Saved {result.tokens_saved:,} tokens")
✓ Saved 2,847,320 tokens ($8.54) — SurfClaw fee: $0.85
0x
Token Reduction
0%
Gross Margin
$0
Fixed Monthly Cost
0%
of Your Savings Only
How It Works

From zero to savings in four steps

01

Install & Connect

pip install surfclaw. Point it at your directory. SurfClaw runs AST extraction (no LLM needed) and semantic extraction in parallel.

02

Graph is Built

Nodes, edges, Leiden clusters, confidence scores. Your entire corpus compressed to a queryable JSON graph — 71.5× fewer tokens than raw files.

03

Query Anything

Ask natural language questions. SurfClaw traverses the graph instead of sending raw files to your LLM — instant answers, massive savings.

04

We Charge 10% of Savings

Lemon Squeezy usage records are filed automatically. You see your savings in real time. SurfClaw takes exactly 10% of documented savings, nothing more.

Capabilities

Enterprise-grade. Startup-priced.

◈

Any corpus. One graph.

Code, PDFs, Markdown, images, meeting transcripts — SurfClaw extracts every concept and relationship into a single traversable knowledge graph. 25 languages via AST, docs via semantic extraction.

⟁

Serverless. Zero fixed cost.

Runs on Modal — you pay only per invocation, never for idle time. Your infrastructure bill is literally $0 when no one is calling the API.

✦

Incremental. Never rebuilds.

Change one file, patch one subgraph. SurfClaw tracks SHA256 content hashes so unchanged nodes are never re-extracted. Enterprise corpora update in under 1 second.

⬡

All LLMs. One interface.

OpenAI, Anthropic, Google, Meta — every model family is supported. Savings are calculated at the exact per-token price of whichever model your team uses.

Quick Start

Three lines to start saving

▶ Terminal
$ pip install surfclaw

▶ Python
from surfclaw import SurfClawClient

client = SurfClawClient(api_key="sk-sc-...")
result = client.process_directory("./src")

# Read the structured report
print(result.report)

# Query the graph in natural language
q = client.query("Why did we choose gRPC over REST?", result.graph)
print(q.answer)

# See your exact savings
print(result.savings_summary())
# Compression: 71.5x | Tokens saved: 2,847,320 | Savings: $8.54 | Fee: $0.85
Pricing

Pay only for what you save

Flat monthly base + 10% of your documented LLM savings. If SurfClaw doesn't save you money, you don't pay the usage fee.

Starter
$49/month
base fee, billed monthly
  • ✓Full knowledge graph API access
  • ✓All LLM providers supported (OpenAI, Anthropic, etc.)
  • ✓Real-time LLM savings dashboard
  • ★+ 10% of your documented LLM savings (usage fee)
  • ✓You keep 90% of everything saved
  • ✓Up to 100 GB corpus / month
  • ✓Email support
Start saving now →

All plans include a 14-day money-back guarantee. · Powered by Lemon Squeezy — cancel anytime.

Get Started Today

Your AI bill is too high.
We fixed it.

Join companies saving thousands per month on LLM costs. Start with $49/mo, pay 10% of your savings.

Start saving now →Read the docs

Install the SurfClaw SDK

$ pip install surfclaw