Riskstate
@likidodefi
Deterministic risk governance for crypto trading agents. Answers "How much can I risk?" with policy level, max position size, leverage limits, and blocked actions — computed from 9+ real-time data sources (on-chain, derivatives, macro, sentiment). Supports BTC and ETH. One tool:
Overview
What is Riskstate?
Riskstate is a risk governance API designed for autonomous crypto trading agents, providing deterministic policy computations from 9+ real-time data sources.
How to use Riskstate?
Invoke the get_risk_policy tool with the required asset parameter (either "BTC" or "ETH"), and optionally supply wallet_address, protocol ("spark" or "aave"), and include_details to receive a full risk policy and classification response.
Key features of Riskstate
- Five policy levels from BLOCK (survival) to GREEN (expansion)
- Max position size as percentage of portfolio (0–100)
- Leverage limit – maximum allowed multiplier
- Allowed and blocked actions for the agent
- Confidence score reflecting signal agreement and data quality
- Classification: cycle phase, market regime, macro regime, direction
- Auditability with composite score, policy hash, and TTL
Use cases of Riskstate
- A trading agent queries risk policy before opening a new position
- Automated rebalancing based on policy level changes
- Compliance logging via policy hash for audit trails
- Multi-source risk assessment with confidence and subscores
FAQ from Riskstate
What parameters does get_risk_policy require?
The asset parameter (BTC or ETH) is required. Optional parameters include wallet_address, protocol (spark or aave), and include_details.
What does the response contain?
The response includes exposure_policy (policy level, max size, leverage, allowed/blocked actions), classification (cycle phase, market regime, macro regime, direction), and auditability (composite score, confidence score, policy hash, TTL).
What do the policy levels mean?
There are five levels: BLOCK, CAUTIOUS, MODERATE, AGGRESSIVE, and GREEN – representing a spectrum from survival to expansion.
How is the confidence score calculated?
The confidence score is a number between 0 and 1, representing the agreement of multiple signals times data quality.
How often is the policy updated?
The policy has a TTL of 60 seconds, meaning it is cached for that duration before a fresh computation is required.
