Skip to content
/ slash Public

AI agent that negotiates your bills down. Built on Telnyx voice + OpenClaw.

Notifications You must be signed in to change notification settings

a692570/slash

Repository files navigation

Slash 🔪

AI agent that calls your providers and negotiates your bills down. You do nothing. It takes 10%.

Americans overpay $50B+ annually on recurring bills. Slash researches competitor pricing, calls your providers over real phone lines, and negotiates better rates — while you sleep.

🏆 Built for the Autonomous Agents Hackathon SF — Feb 27, 2026 — $47k+ in prizes


How It Works

  1. Add your bills — internet, phone, insurance, subscriptions
  2. Slash researches — finds competitor offers, builds your leverage profile
  3. Slash calls — dials your provider's retention line and negotiates live
  4. You save — average $240/year per bill. We take 10% of savings.

Architecture

┌─────────────┐     ┌──────────────────┐     ┌─────────────────┐
│   User UI   │────▶│   Slash Server   │────▶│  Telnyx Voice   │
│  (Render)   │     │   (Node.js)      │     │  AI Assistants  │
└─────────────┘     └──────┬───────────┘     └────────┬────────┘
                           │                          │
                    ┌──────┴───────┐           ┌──────┴────────┐
                    │              │           │               │
              ┌─────▼─────┐ ┌─────▼─────┐  ┌──▼───┐   ┌──────▼──────┐
              │  Tavily    │ │  Yutori   │  │ PSTN │   │  Modulate   │
              │  Research  │ │  Browse   │  │ Call │   │  Velma      │
              └─────┬──────┘ └───────────┘  └──────┘   └─────────────┘
                    │                                         
              ┌─────▼─────┐  ┌───────────┐  ┌──────────────┐
              │   Neo4j   │  │    AWS    │  │   Senso     │
              │   Graph   │  │  S3/logs  │  │  Playbooks  │
              └───────────┘  └───────────┘  └──────────────┘

Sponsor Integration Strategy

We integrate 9+ sponsor APIs/products to maximize prize track eligibility:

# Sponsor/Tech Integration Notes
1 Telnyx Call Control + AI Assistants for real outbound negotiation calls Core voice infrastructure
2 AWS Call recordings stored in S3, Bedrock for analysis Cloud infra
3 OpenAI GPT-4o powers the negotiation AI brain (via Telnyx AI Assistants) Core AI
4 Render Production deployment platform 2 judges from Render
5 Tavily Competitor pricing research API — real-time market data Already integrated
6 Yutori Browsing API for deep web research on provider sites, account verification, bill scraping 1 judge (chief scientist)
7 Neo4j Knowledge graph storing provider data, negotiation outcomes, competitor rates Graph DB
8 Modulate Velma voice intelligence monitors calls for quality, safety, emotion detection 2 judges (CTO + COO)
9 Senso Knowledge management for negotiation playbooks and best practices KM angle
10 Numeric Financial data analysis and bill tracking Finance angle

Tech Stack

  • Runtime: Node.js + TypeScript
  • Voice: Telnyx Call Control + AI Assistants (GPT-4o)
  • Research: Tavily API + Yutori Browsing + Research API
  • Database: Neo4j (knowledge graph), AWS S3 (recordings)
  • AI: OpenAI GPT-4o (via Telnyx), AWS Bedrock
  • Voice Intelligence: Modulate Velma
  • Knowledge Base: Senso
  • Deployment: Render
  • Agent Framework: OpenClaw (optional skill mode)

Quick Start

npm install
cp .env.example .env  # add your API keys
npm run dev            # dashboard at localhost:3100

Required API Keys

TELNYX_API_KEY=
OPENAI_API_KEY=
TAVILY_API_KEY=
NEO4J_URI=
NEO4J_PASSWORD=
YUTORI_API_KEY=
MODULATE_API_KEY=
AWS_ACCESS_KEY_ID=
AWS_SECRET_ACCESS_KEY=

OpenClaw Skill Mode

openclaw skill install slash
# Slash runs autonomously — weekly bill checks, auto-negotiation, memory logging

Team

Built by Abhishek at the Autonomous Agents Hackathon SF — Feb 27, 2026

Slash saves you money so you don't have to sit on hold.

About

AI agent that negotiates your bills down. Built on Telnyx voice + OpenClaw.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages