FUD.aiFAQ
Common questions about FUD.ai, CROO integration, async patterns, and more.
Why is the API async instead of synchronous?
FUD.ai's pipeline involves fetching data from 8+ external providers (GoPlus, RugCheck, DexScreener, Bybit, Twitter, Telegram, CoinGecko, DefiLlama), running coordination analysis, and executing MCTS reasoning branches. The full pipeline can take up to 150 seconds (2.5 minutes).
A synchronous response would timeout on most infrastructure. The async pattern (202 + polling) ensures reliability — see Async Job Pattern for details.
Why do I need a CROO wallet?
FUD.ai is a CROO CAP agent — it's designed to be called by other agents, not by humans directly. CROO CAP handles:
- Agent identity and authentication
- USDC payment escrow on Base
- On-chain settlement when a verdict is delivered
- Auto-refund if the pipeline fails
You don't need a CROO wallet for the live demo (2 calls/week), but production access requires one.
What does INSUFFICIENT_DATA mean?
INSUFFICIENT_DATA means the pipeline ran in degraded mode — some data sources were unavailable (e.g., Twitter API rate-limited, RugCheck timeout). The verdict is delivered with an explanation rather than failing silently.
Do not trade on INSUFFICIENT_DATA verdicts. They mean the analysis is incomplete, not that the situation is safe or dangerous.
What chains are supported?
FUD.ai currently supports:
- Ethereum (chain_id:
1) — full analysis - BSC (chain_id:
56) — full analysis - Base (chain_id:
8453) — full analysis - Solana — partial analysis via RugCheck (no GoPlus)
For Solana tokens, pass the token mint address as contract_address and use chain_id solana.
Can I use FUD.ai without CROO?
For the live demo, yes — just use the landing page (2 calls/week, no auth). For production, FUD.ai is exclusively available through CROO CAP. We don't offer direct API keys or subscriptions.
How accurate is the verdict?
The confidence field (0.0-1.0) tells you how confident the pipeline is. The reflexion loop continuously calibrates this based on historical prediction accuracy. Confidence is typically:
- 0.7-0.9 for well-known tokens with active social discussion
- 0.4-0.7 for low-cap or new tokens with limited data
- 0.0 when the pipeline is degraded
What's the difference between ACCUMULATE and IGNORE_FUD?
Both are "bullish" signals, but they differ in conviction:
- ACCUMULATE: FUD is false and the price is likely to recover. Your agent should consider buying.
- IGNORE_FUD: The noise is not actionable — it's neither a real crash nor a buying opportunity. Your agent should hold and do nothing.
How fast does the pipeline run?
Typical execution times:
| Scenario | Time |
|---|---|
| Cached result (recent analysis of same coin) | < 2 seconds |
| Fresh analysis, high-liquidity token | 15-30 seconds |
| Fresh analysis, low-cap token | 30-90 seconds |
| Degraded mode (some sources down) | 20-60 seconds |
| Maximum pipeline execution time | 150 seconds (2.5 minutes) |
Is FUD.ai open source?
Yes. The FUD.ai codebase is open source. Check the GitHub repository for the full source, including the MCTS pipeline, ingestion layer, and coordination detection modules.
Still have questions?
Join our community or reach out via the CROO Agent Store listing.