Author: Guavy Inc. Contact:donna@guavy.com Date: Donna Tilden, CPA, CFA
Summary
Guavy is an AI-powered analytics platform that ingests and analyzes between 500,000 and 1,500,000 off-chain sources daily, generating real-time sentiment and signal analytics for over 500 cryptocurrencies. We propose to provide Aave’s Risk DAO with a free pilot feed of our off-chain sentiment indicators to evaluate its potential value in enhancing Aave’s risk management strategies.
Motivation
While Aave already benefits from robust on-chain data and market metrics, key trends and risks often originate in off-chain domains (e.g., regulatory developments, social media volatility, reputational events, etc.). These events frequently precede on-chain movements. Guavy’s dataset fills this gap by:
Quantifying market mood from thousands of financial news outlets, forums, and social platforms
Identifying shifts in sentiment before on-chain actions reflect them
Providing signal data aligned with professional heuristics
We believe this data could support:
Liquidity risk forecasting
Collateral risk profiling
Volatility anticipation and stress testing
Proposed Pilot for Risk DAO
We are offering the Risk DAO access to a free 30-day pilot feed covering a subset of Aave-listed assets. This includes:
Daily sentiment score with trend direction (1D, 7D, 30D views)
Signal alerts (where relevant) based on sentiment inflection points
Coin coverage: ETH, WBTC, AAVE, USDC, USDT, DAI
Delivery method: API and/or daily dashboard snapshot
Evaluation goal: Determine if sentiment data enhances risk decisioning (ex-ante signal correlation, reputational triggers, etc.)
Request to the Community
We are seeking feedback from Risk DAO members on:
Interest in trialing this data
Suggestions for additional assets or features
Preferred format for data delivery
We welcome the opportunity to collaborate and co-design this evaluation with the DAO and are happy to tailor the scope or timeline as needed.
Me personally, Im not sure how Guavy could benefit the DAO with risk decisions, as its more or less a sentiment checker (at least thats what I understood). It feels like its more of a tool for trader to gauge sentiment for long or short positions.
Totally fair point, and I appreciate the skepticism — Guavy is indeed sentiment-driven, but it’s more than just a “long/short” trader’s tool.
Here’s how Guavy could support Aave DAO on the risk side:
Off-chain sentiment as an early risk signal
We ingest and analyze 500k–1.5M off-chain sources daily (e.g., news, forums, Twitter/X, blogs) across 500+ assets. Sudden shifts in sentiment often precede on-chain activity — especially panic withdrawals or liquidations triggered by rumors, exchange outages, or macro news. That makes Guavy useful for proactive alerting on asset-specific or systemic stress before it hits protocol metrics.
Sentiment volatility as a stability input
We provide asset-level sentiment volatility charts and signal alerts. This can help protocols like Aave assess market mood fragility or gauge how reactive certain communities are — useful for risk scoring collateral assets or tweaking parameters like LT or RF.
Supplement to risk parameter modeling
We’re not suggesting replacing Chaos or Llama’s models — just adding a signal layer. For example, if an asset like CRV is facing bearish sentiment across Reddit, Twitter, and media but no on-chain signals yet, it might warrant early monitoring or a heads-up.
If helpful, happy to put together a one-pager or live demo for the DAO’s risk contributors. Guavy isn’t a silver bullet, but it’s a complementary risk intelligence tool that’s cheap to test and easy to integrate.