SiftrCode v0.1.1

Cut agent token waste by 90%

Our proprietary synthesis engine strips implementation bloat into clean interfaces before inference.

Demo:
user_service.ts
scanning
Step 1: Raw Code
0 tokens $0.00
saved 0 tokens $0.00

Works with Claude Code & Cursor

Select your coding agent to view instant setup commands.

Connect to Claude Code

• Guide →

Registers SiftrCode's Stdio MCP server to prune code before Claude reads it.

$ claude mcp add siftrcode -- npx -y siftrcode mcp

Token Economics

The math is easy. You could be saving thousands in agent fees*

Context Window

Payload per Turn

−92% Tokens
125k
62k
125,000
tokens (100%)
Raw
−92.2%
9,800
tokens
AST
Without Siftr
With SiftrCode

Method bodies dissolved • 100% signatures & types preserved.

Claude Spend

Monthly Cost / Dev

Save $380 / mo
$412
$200
$412.50/mo
$4,950 / yr
$0.38/turn
Save $380
$32.34/mo
$388 / yr
$0.03/turn
Without Siftr
With SiftrCode

Saves $0.35/turn • $4,560 net annual savings per engineer.

Net ROI
+$4,560/ dev
Latency
11xfaster
10-Dev Team
+$45.6k/ yr
$ npx siftrcode audit

* Analysis for Claude 3.7 Sonnet ($3.00/MTok in) based on 50 agent turns / day.

Simple Licensing

Free open-source CLI for individuals. Paid caching for teams.

OPEN SOURCE
INDIVIDUAL DEVELOPERS

Community CLI

$0 Free forever (MIT)
  • ✔ Local AST Pruning (TS, Py, Go, Rust)
  • ✔ Codebase Auditor ($siftrcode audit)
  • ✔ Context Packer ($siftrcode pack)
  • ✔ Stdio MCP Server integration
GitHub Repository →
TEAM PRO
FOR STARTUPS & TEAMS

SiftrCode Pro

$19 / seat / mo
  • ✔ Everything in Community CLI
  • ✔ TypeSafe Jev Cloud Relevance Gate
  • ✔ Team Cross-Repo Context Cache
  • ✔ CI/CD Pull Request Context Auditor