Security is a continuous process. As AI agents have become powerful at spotting and exploiting bugs in smart contract at scale, protocols are at a natural disadvantage. This means that an independent audit is not sufficient, and continuous audits are highly desirable. While attackers use AI to gain an edge in exploiting bugs, whitehat hackers are taking the opposite side, helping projects defend from them. That is why we have a standing bug bounty program. Since the program opened, we have received over 80 security reports, 10 of which have been confirmed and rewarded. We are grateful to the community of researchers who keep looking. Building in the open gives attackers an edge; a proactive security process is how we fix issues before that edge is used.
On top of documenting findings, we keep the known-issues page current so researchers do not spend time on design decisions we have already accepted. We try to respond quickly, but triage is by impact and likelihood, not by arrival order. A report that looks like an imminent loss of funds can be confirmed in hours. Another can sit for days or weeks. Bandwidth is limited, and imminent threats come first. During busy periods some reports get no further reply after a first look. That is normal: every report gets a preliminary read, and anything that looks like a false positive or a documented known issue drops down the queue.
The bounty is the public layer of that process. Rigoblock V4 has been in scope since 29 May 2025. V3 is deprecated and is not covered. Critical issues that can take smart-pool funds pay up to $10,000, high up to $2,000, medium up to $250, in stablecoin, and high and critical payouts are also capped at 10% of protocol TVL at risk, with minimum amounts to discourage withholding the finding.
A lot of plausible-looking reports are out of scope on purpose. The relationship between a pool operator and its holders is trust-minimized, not trustless. The operator can set parameters, choose tokens that have a price feed, and route trades. Attacks that need a malicious or compromised operator, a governance takeover, oracle manipulation, rounding, or a third-party contract we do not control are not paid. The same is true of the cases written up under known issues: front-running by an operator, rogue tokens the operator has approved, cross-chain sync lag, Uniswap v4 hooks, and accepted NAV understatements or net virtual supply mismatch. Those pages exist so a report on them can be closed as informational.
Reports go only to security@rigoblock.com, within 24 hours of discovery, and stay private until we have fixed the issue and allowed disclosure. A useful report says how to reproduce the bug, has a proof of concept on a local fork, assesses the severity and what the impact is. Live testing on public mainnet or testnet is not allowed. Researchers may be credited after the fix.
The practical reading is narrow: do not treat the operator as a trustless party. Do not expect the bounty to cover operator misbehavior, oracle games, or issues we have already listed. If you are looking for bugs, read the program and the known-issues page before you write the report. The rules live at docs.rigoblock.com/bug-bounty.

